Preliminary study on the short-term changes of pulmonary perfusion after a single balloon pulmonary angioplasty in patients with chronic thromboembolic pulmonary hypertension

Abstract

Background:

Little is known about immediate responses of blood perfusion to the balloon pulmonary angioplasty (BPA) procedure.

Objectives:

To investigate the changes in pulmonary perfusion of balloon-dilated vessels and untreated vessels with before, immediately after a single BPA and at follow-up.

Design:

Retrospective single-center cohort study.

Methods:

Patients who had chronic thromboembolic pulmonary hypertension (CTEPH) and completed the pulmonary perfusion single photon emission computed tomography (SPECT) imaging before, immediately after BPA and at follow-up were included. We evaluated the perfusion defects of both-lung, BPA target (balloon dilated) and non-target (untreated) vessel segments according to Begic 3-point scale in each lung segment.

Results:

Forty patients (40 BPA procedures) were included and were given next BPA after 89 (62–125) days. The hemodynamic parameters including mPAP, PVR, and RAP were significantly improved after a single BPA. Visual scoring results of pulmonary perfusion imaging in 40 BPAs showed the perfusion defect scores of target vessels reduced from 5.6 ± 2.6 to 4.2 ± 2.2 (p < 0.001) immediately after BPA, and then further diminished to 3.1 ± 1.9 (p < 0.001) at follow-up. While in the non-target vessels, the post-BPA perfusion defect scores did not change significantly (13.4 ± 4.7 versus 12.8 ± 4.6, p = 0.182), but tended to decrease at follow-up (12.2 ± 4.2). However, there were 17 BPAs of which the post-BPA perfusion defect scores of non-target vessels increased significantly (p < 0.001), but decreased at follow-up.

Conclusion:

In addition to improving the blood perfusion of target vessels, BPA also has a certain effect on the perfusion of some non-target vessels.

Introduction

Chronic thromboembolic pulmonary hypertension (CTEPH) is a subtype of pulmonary hypertension classified as Group 4. ¹ It is characterized by stenoses or obstructions of the pulmonary arteries by residual organized thrombi and small-vessel arteriopathy relevant to high shear stress in patent vessels, causing elevated pulmonary vascular resistance (PVR), progressive PH, ultimately leading to right heart failure. CTEPH is a rare, progressive pulmonary vascular disease that seriously endangers human health. Without effective treatment, the prognosis of CTEPH is poor, especially in patients whose mean pulmonary arterial pressure (mPAP) exceeds 30 mmHg. ² The potentially curative treatment for CTEPH is pulmonary endarterectomy (PEA). ³ However, around 40% of patients with CTEPH are ineligible for PEA. ⁴ Balloon pulmonary angioplasty (BPA) is an emerging percutaneous revascularization for CTEPH patients who are unable to undergo PEA or still have PH after PEA.^5,6 In the past decade, advances in the techniques of BPA have led to better clinical outcomes with improvements in pulmonary hemodynamics, pulmonary perfusion, symptoms, exercise capacity, and quality of life.^1,3 Both PEA and BPA improve the abnormal hemodynamics of CTEPH. However, there are some differences between these treatments. Previous studies have found that PEA can cause redistribution of blood perfusion in pulmonary artery.^7–9 BPA dilates multiple thrombotic lesions, dramatically improving the peripheral perfusion and PVR. On the other hand, BPA can also theoretically relieve the excessive flow and pressure in the non-BPA vascular area, thus affecting blood perfusion in the non-BPA vascular area, but the blood perfusion impact of BPA on the balloon-dilated and untreated vascular area has not been extensively studied.

In this study, we used perfusion single photon emission computed tomography (SPECT) imaging to clarify the change of blood perfusion in the BPA and non-BPA area before and after BPA, understanding the mechanism of BPA treating CTEPH.

Methods

The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement (Supplemental File).

Study design

This retrospective study included CTEPH patients in our hospital from 1 June 2019 to 31 August 2020. We screened patients for eligibility of BPA and identified patients who completed pulmonary perfusion imaging before BPA (within 3 days), after BPA (within 3 days) and right heart catheterization (RHC) before and after balloon dilation. And follow-up pulmonary perfusion SPECT images and RHC were also performed within 1 week of next-BPA session. The study was reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Guangzhou Medical University. Informed consent was waived due to the retrospective design and minimal risk nature of the study.

After applying these selection criteria, a total of 40 patients (40 BPAs) were included in this study. Time intervals between BPA and follow-up pulmonary perfusion imaging (days) were determined from the medical records. Pulmonary vasodilator therapy and unspecific treatments such as oral anticoagulants, diuretics, digitalis or oxygen supplementation are permitted, but the treatment needs to be stable for at least one month before BPA and unchanged until the follow-up pulmonary perfusion SPECT images and RHC.

Pulmonary perfusion SPECT/CT images acquisition

Imaging was performed using a Siemens Symbia T16 SPECT/CT system (Siemens Healthcare, Munich, Germany) with a low energy high resolution collimator and Zoom of 1.0. Patients were allowed to breathe steadily in the supine position during scanning. After injection of a median dose of 140 (25th–75th percentile range, 132-161) MBq ^99mTc-MAA (Guangzhou Atom High-tech Isotope Pharmaceutical Co., Ltd., Guangzhou, China), SPECT scanning was performed in a 128 × 128 matrix over 360° (with a 6° step and 15 s per frame). A low-dose CT was then performed in the same position (3-mm slice thickness; 120 kV) under free breathing. SPECT/CT were obtained simultaneously within 10 min. Reconstruction was performed using an iterative ordered-subsets expectation maximization algorithm (8 iterations, 16 subsets) and processed with Butterworth postfiltering (power 10.0). And the attenuation-corrected SPECT data were fused with the CT (2-mm reconstruction slice thickness).

Perfusion SPECT/CT images analysis

SPECT/CT images were interpreted on a Siemens syngo MI workstation, which allowed the simultaneous review of SPECT, CT, and fused imaging data. All images were interpreted independently by two trained observers (S-Y.L. and P.H., with 5 and 8 years of cardio-pulmonary nuclear medicine experience, respectively) without knowledge of the subjects’ histories. A third investigator (X-L.W., with 20 years of cardio-pulmonary nuclear medicine experience) acted as a mediator to resolve any discrepancies. Pre- and post-BPA SPECT/CT images were analyzed in a random order and at different times to avoid memory bias. The images were analyzed using visual scoring of perfusion defect extent with a semiquantitative scoring system. For semiquantitative scoring, perfusion SPECT images were analyzed according to Begic 3-point scale ¹⁰ (0, normal; 1, subsegmental defect or segmental reduction; 2, segmental defect; Figure 1). The pulmonary arteries were subdivided into 18 pulmonary artery segments: 10 right lung and 8 left lung per patient. Each lung segment was scored separately on SPECT/CT images, and the individual scores for all segments were then summed up, yielding the perfusion defect score for the both-lung, BPA target (balloon dilated) and non-target (untreated) vessel segments.

Figure 1.

Visual score map of perfusion defect extent at lung segment level. For semiquantitative scoring, perfusion SPECT/CT images were analyzed using a three-point scale (0, normal; 1, subsegmental defect or segmental reduction; 2, segmental defect) in axial, sagittal, and coronal planes (left to right); (a) No perfusion defect (score 0) was evident in the Left S6 (dotted line); (b) a subsegmental defect in the left S8 (score 1) (dotted line); (c) a segmental perfusion defect in the Right S1 (score 2) (dotted line).

Pulmonary hemodynamics studies

RHC was performed at the beginning of BPA (pre-BPA), at the end of session (post-BPA) and at the follow-up period (follow) of BPA. Hemodynamic parameters, including pulmonary artery systolic pressure (sPAP), pulmonary arterial diastolic pressure (dPAP), mPAP, pulmonary arterial wedge pressure (PAWP), right atrial pressure (RAP), cardiac output (CO), cardiac index (CI) and PVR were measured. CO was determined by the indirect Fick method and corrected for body surface area (CI).

BPA procedure

BPA was performed as a staged procedure, with treatment of a limited number of pulmonary segments during each session. The standard procedure has been described previously. ¹¹ Briefly, after RHC with intravenous heparin, a selective pulmonary angiography was performed to evaluate the characteristics of the lesions. We selected targeted vessels based on comprehensive findings, including webs, bands, abrupt narrowing, and complete obstructions, obtained by pulmonary angiography, DynaCT angiographic reconstruction and optical computed tomography. A 0.014-inch guidewire was passed through the target lesion, and the target lesion was dilated to an appropriate size by multiple balloon inflations manually using balloon catheters depending on vessel diameter. After inflation, contrast agent was injected into the treated vessel to evaluate the angiographic effect of the procedure. We treated 2–10 segmental or subsegmental arteries in each procedure session according to patient severity, time of procedure (not exceed 5 h), and the amount of contrast media given.

Statistical analysis

Categorical variables were presented as count (percentage), continuous variables were expressed as mean ± standard deviation (SD) or median [interquartile range (IQR), 25th and 75th percentile]. Shapiro–Wilk tests were performed to test the normality of data. If the Shapiro-Wilk test revealed normal distribution, one-way repeated measures analysis of variance (ANOVA) was applied to compare the change of vessel scores and hemodynamic parameters before BPA, after BPA, and at follow-up. Otherwise, the non-parametric Friedman test was used. When the ANOVA or Friedman test revealed significance, we performed Bonferroni’s multiple comparisons test. Kappa (κ)-values to determine the interobserver variability between nuclear medicine physician readers were assessed. A p-value <0.05 was taken to indicate statistical significance. Statistical analyses were conducted using SPSS statistics 25.0 (IBM, Armonk, NY, USA).

Results

Patient baseline characteristics

Table 1 shows the baseline patient characteristics. The predominance of elderly subjects (60 ± 9 years) and female gender reflected the epidemiology of CTEPH in China. The mPAP was 43.2 ± 12.5 mmHg, PVR was 9.8 ± 5.4 Wood units, CI was 2.5 ± 1.1 L/min/m², pro-brain natriuretic peptide (pro-BNP) was 299.8 (91.0–1539.0) pg/mL, and 6-minute walk distance (6-MWD) was 382.5 ± 114.4 m, indicating that most participants had mild to moderate disease severity at baseline.

Table 1.

Baseline characteristics (40 patients).

Age	Years	60 ± 9
Gender	Male/female, n	6/34
WHO functional class	I/II/III/IV, n	0/25/13/2
Usage of pulmonary vasodilator	PDE5i, n (%)	10
	sGC, n (%)	15
	ERA, n (%)	9
	Prostanoid, n (%)	5
Anticoagulation	n (%)	40 (100%)
mPAP	mmHg	43.2 ± 12.5
CI	L/min/m2	2.5 ± 1.1
PVR	Wood units	9.8 ± 5.4
pro-BNP	pg/mL	299.8(91.0–1539.0)
6-MWT	m	382.5 ± 114.4

CI, cardiac index; ERA, endothelin receptor antagonist; 6-MWT, six-minute walk distance; mPAP, mean pulmonary arterial pressure; PDE5i, phosphodiesterase type 5 inhibitor; pro-BNP, pro-brain natriuretic peptide; PVR, pulmonary vascular resistance; sGC, soluble guanylyl cyclase stimulator.

Pulmonary hemodynamic changes after a single BPA

The BPA procedure was well tolerated in all patients. Two patients had a small amount of hemoptysis during balloon dilatation. Cessation of the operation and compression of the damaged artery with balloon for an average of 10 min resulted in the disappearance of hemoptysis without additional medications or rescue therapy. And no other BPA-related complications were observed. We evaluated the findings from 40 BPA procedures (40 patients) with a median follow-up of 89 (62–125) days. Table 2 and Figure 2 summarize the changes of hemodynamic parameters including mPAP, PVR, and RAP before and after BPA. A single BPA session decreased the mPAP from 43.2 ± 12.5 to 36.4 ± 12.0 mmHg (p < 0.001), PVR from 9.8 ± 5.4 to 7.5 ± 4.1 Wood units (p < 0.001) and RAP from 7.5 ± 4.2 to 6.1 ± 3.2 mmHg (p = 0.014). All patients maintained the original anticoagulant and PH targeted drug treatment after BPA.

Table 2.

Changes of visual score of SPECT/CT in pulmonary perfusion.

Variables	Pre-BPA	Post-BPA	Follow-up	F ratio	p
Visual score
Whole lung	18.5 ± 4.9	17.7 ± 4.7	15.3 ± 4.1	24.116	<0.001
Target vessels	5.6 ± 2.6	4.2 ± 2.2	3.1 ± 1.9	37.381	<0.001
Non-target vessels	12.8 ± 4.6	13.4 ± 4.7	12.2 ± 4.2	8.459	<0.001
Hemodynamic
sPAP (mmHg)	75.3 ± 24.3	70.4 ± 22.9	64.0 ± 22.9	15.197	<0.001
dPAP (mmHg)	24.5 ± 7.8	22.6 ± 7.3	21.1 ± 7.6	8.705	<0.001
mPAP (mmHg)	43.2 ± 12.5	40.0 ± 11.8	36.4 ± 12.0	24.766	<0.001
PAWP (mmHg)	9.4 ± 3.1	9.5 ± 3.2	9.2 ± 3.1	0.055	0.826
RAP (mmHg)	7.5 ± 4.2	7.1 ± 4.6	6.1 ± 3.2	5.367	0.015
CI (L/min/m2)	2.5 ± 1.1	2.7 ± 0.9	2.7 ± 0.9	0.556	0.576
CO (L/min)	4.1 ± 1.7	4.4 ± 1.7	4.3 ± 1.5	0.559	0.575
PVR (Wood units)	9.8 ± 5.4	8.4 ± 4.8	7.5 ± 4.1	10.563	<0.001
pro-BNP (pg/mL)	299.8 (91.0–1539.0)	307.2 (113.4–1191.0)	110.4 (60.3–373.2)	24.051	<0.001*

Data are compared using one-way repeated measures analysis of variance (ANOVA) or Friedman test respectively.

^*Data were analyzed using the non-parametric Friedman test.

BPA target vessels, balloon-dilated segments; CI, cardiac index; CO, cardiac output; dPAP, pulmonary arterial diastolic pressure; mPAP, mean pulmonary arterial pressure; Non-target vessels, untreated segments; PAWP, pulmonary arterial wedge pressure; pro-BNP, pro-brain natriuretic peptide, PVR, pulmonary vascular resistance; RAP, right atrial pressure; sPAP, pulmonary artery systolic pressure.

Figure 2.

Changes of visual scores in pulmonary perfusion and pulmonary hemodynamic after a single BPA.

Visual scoring results showed that compared with pre-BPA, the post-BPA perfusion defect scores of the whole lung did not change significantly (p = 0.415), but improved significantly during follow-up period (p < 0.001). In target (balloon dilated) segments, the perfusion defect scores reduced immediately after balloon dilation (p = 0.001), and then further decreased at follow-up period (p < 0.001). While in the non-treated segments, the post-BPA perfusion defect scores did not change significantly immediately after balloon dilation (p = 0.182), but tended to decrease at follow-up period (p = 0.182). And after a single BPA session, hemodynamic parameters including mPAP, PVR and RAP were decreased significantly.

BPA, Balloon pulmonary angioplasty; mPAP, mean pulmonary arterial pressure; PVR, pulmonary vascular resistance; RAP right atrial pressure.

Visual scores changes in pulmonary perfusion after a single BPA

The κ-analyses indicated that the results of visual scores interpreted by two nuclear medicine physicians were in good agreement (κ = 0.86). The changes in visual scores of pulmonary perfusion after a single BPA were evaluated in 40 patients with CTEPH (Table 2 and Figure 3). For the whole lung, the post-BPA perfusion defect scores did not change significantly compared with the baseline at pre-BPA (17.7 ± 4.7 versus 18.5 ± 4.9, p = 0.415), but improved significantly during follow-up period (15.3 ± 4.1 versus 18.5 ± 4.9, p < 0.001). In target (balloon dilated) segments, the perfusion defect scores reduced from 5.6 ± 2.6 to 4.2 ± 2.2 (P = 0.001) immediately after balloon dilation, and then further diminished to 3.1 ± 1.9 (P < 0.001) at follow-up period. While in the non-target (untreated) segments, the post-BPA perfusion defect scores did not change significantly immediately after balloon dilation (13.4 ± 4.7 versus 12.8 ± 4.6, p = 0.182), but tended to decrease at follow-up period (12.2 ± 4.2).

Figure 3.

The change of pulmonary perfusion in deterioration group and non-deterioration group.

In deterioration group, namely, the non-target blood vessel score increased after BPA, there were 17 BPAs observed a decrease of perfusion defect scores in the target segments after BPA without statistical difference, but an increase of perfusion defect scores in the non-target segments (p < 0.001). And in the non-deterioration group, both target and non-target segments perfusion scores of 23 BPAs decreased immediately after BPA and then further diminished at follow-up period.

BPA, Balloon pulmonary angioplasty.

According to whether the non-target blood vessel score increased after BPA (Figure 4), 40 BPAs were divided into deterioration group (17 BPAs) and non-deterioration group (23 BPAs). In non-deterioration group, both target and non-target segments perfusion scores decreased immediately after balloon dilation and then further diminished at follow-up period. In deterioration group, the perfusion defect scores in target segments tended to reduced immediately after balloon dilation (p = 0.281), and then further diminished at follow-up period (p < 0.001). While in the non-target segments, compared with the pre-BPA, the post-BPA perfusion defect scores increased significantly immediately after balloon dilation (p < 0.001), but decreased at follow-up period.

Figure 4.

Pulmonary perfusion SPECT/CT images of a 43-year-old female with CTEPH before BPA, after BPA, and during follow-up period. (a) The left S8 (target vessel) showed subsegmental defect before BPA. The distribution of perfusion improved significantly after BPA and during follow-up period (visual score: 1-0-0) (red arrow); (b) The left S9 (non-target vessel) was generally normal before BPA, and the distribution of perfusion showed a subsegmental defect after BPA, and the radioactive distribution returned to normal during follow-up period (visual score: 0-1-0) (white arrow), taking into account the phenomenon of ‘steal blood’; (c) The right S8 (non-target vascular region) showed a subsegmental defect before and after BPA, and the distribution of perfusion improved during follow-up period (visual score: 1-1-0) (yellow arrow).

BPA, Balloon pulmonary angioplasty.

Additional information regarding baseline characteristics and hemodynamic data of deterioration group and non-deterioration groups is specified in the Supplemental Tables 1 and 2.

Visual scores changes based on segments in pulmonary perfusion after a single BPA

Figure 5 shows the changes in visual scores of 720 pulmonary artery segments in 40 BPAs at three different time points. For target segments (n = 211), scores of baseline were 0 in 39 segments (18.5%), 1 in 120 segments (56.9%), and 2 in 52 segments (24.6%). After BPA, scores were 0 in 63 segments (29.9%), 1 in 128 segments (60.7%), and 2 in 20 segments (9.5%). At follow-up, scores were 0 in 98 segments (46.4%), 1 in 106 segments (50.2%), and 2 in 7 segments (3.3%). For non-target segments (n = 509), the visual score of 41 segments increased after BPA and 7 segments increased at follow-up, indicating the blood perfusion decreased. While the visual score decreased in 37 lung segments after BPA and/or at the follow-up period and did not change in 424 segments.

Figure 5.

Changes of visual scores in 720 pulmonary artery segments (40 BPAs) at three different time points.

BPA, Balloon pulmonary angioplasty.

Discussion

The mechanism of pulmonary hypertension in CTEPH is multifactorial. Recent insights have revealed that the important role of hemodynamic stress in the progression of small vessel arteriopathy in CTEPH. In CTEPH, non-occlusive pulmonary arteries exposed to high pressure and shear stress undergo redistribution of pulmonary blood flow, leading to endothelial dysfunction, inflammation and abnormal cell proliferation and resulting in vascular damage similar to those observed in idiopathic pulmonary hypertension. These changes lead to a progressive increase of PVR, which eventually develops into symptomatic CTEPH. ¹² From this perspective, BPA, an interventional therapy developed in recent years, mechanically dilates the stenotic/occluded pulmonary arteries with a balloon catheter to restoring blood flow and reducing hemodynamic stress, that may slow the progress of vascular remodeling.

In order to reduce complications such as hemoptysis and reperfusion pulmonary edema, BPA is performed as a staged procedure, with treatment of a limited number of pulmonary segments during each session.^6,13 Our study showed that there was an improvement of hemodynamics after a single BPA session, manifested by the decrease of PAP and PVR. However, CI did not show any significant changes but presented a rising trend after a single BPA session among our patients, which might be ascribed to a single session and small sample size of our study. Although the changes of pulmonary hemodynamics after BPA have been reported, the focus on the changes of blood perfusion including target (balloon dilated) and non-target (untreated) vessels after BPA has not been fully studied.

At present, pulmonary perfusion SPECT/CT imaging is one of the noninvasive tools for quantitative assessment of pulmonary perfusion defect. SPECT/CT is based on microembolization of pulmonary capillaries and precapillary arterioles, without visualization of bronchial arterial flow. Furthermore, the SPECT/CT technique is robust and well standardized, therefore, the lung perfusion SPECT/CT is an ideal tool for assessing pulmonary perfusion. ¹⁴ In this study, we first used pulmonary perfusion SPECT/CT fusion imaging to quantitatively evaluate the degree of impaired pulmonary blood flow perfusion and treatment response in patients with CTEPH after a single BPA. We studied the pulmonary perfusion changes using visual score (Begic method) in the BPA and non-BPA area before and after BPA. In our case series, we found that compared with pre-BPA, although there was no statistical difference in the pulmonary perfusion score of bilateral lungs in all patients immediately after BPA, the blood perfusion of bilateral lungs was further improved at follow-up (p < 0.001). Similar hemodynamic changes were also observed after BPA,^15,16 suggesting that the improvement of blood perfusion after BPA is more significant over time.

In order to further investigate the causes of this phenomenon, we analyzed the perfusion visual score of target vessels and non-target vessels. After balloon dilatation, we found that pulmonary perfusion in the BPA area of all patients improved, that is, the visual score decreased, and further improved over time. The possible reasons for the analysis are as follows: After BPA, the fibrous network structure in the lumen of the target vessel was dilated by the balloon, and the lumen expanded, so that the vascular resistance of the target vessel decreased and blood flow perfusion improved. At the same time, after the blood flow in this area increased, anticoagulant drugs can enter more into the lung segments downstream from partially occluded arteries to better prevent in situ thrombosis. ¹⁷

On the other hand, we found that the blood perfusion of non-target vessels after BPA had a tendency to deteriorate, although there was no statistical difference. We further found that there were 17 BPA procedures (42.5%) of which the blood perfusion in the non-target vessel decreased in post-BPA perfusion scans. The complete absence of perfusion in previously perfused lobar or segmental regions or the appearance of new defects subsequent to the post-BPA perfusion scans should raise suspicion for recurrent embolic events. However, considering that these patients were receiving anticoagulant therapy and had no corresponding clinical symptoms, we excluded acute pulmonary embolism and speculated that the change of blood perfusion in the non-target vascular area after BPA may be due to the phenomenon of ‘steal blood’. That is, the blood flow of the target vessels increased, but the blood flow in the untreated vessels decreased. After 89 (62–125) days, the reexamination of pulmonary perfusion showed that perfusion defects in non-target vessel areas were gradually improved (Figure 3).

In CTEPH, increased blood flow and wall shear stress over time cause adaptive changes and increased pulmonary vasculature resistance in ‘open’ vascular bed that were not affected by pulmonary embolism. We speculate that such hypertensive changes in the vascular bed that was ‘open’ preoperatively may reduce flow to these zones after BPA, so the pulmonary artery blood flow is shunted to the target vessel area where the PVR decreases after balloon dilatation, and the blood flow redistribution occurs. Hosokawa et al. ¹⁸ combined with phase contrast magnetic resonance imaging (PC-MRI) to provide quantitative local blood flow and RHC, to clarify changes in blood flow, pressure, and PVR in the non-BPA area before/after BPA. The results also confirmed our hypothesis that BPA reduced PVR and total pulmonary artery pressure on the BPA-side, thereby diverting pulmonary blood flow from the non-BPA-side to the BPA-side. This phenomenon has been reported in patients with CTEPH after PEA.^7–9 It has been reported that hemodynamics improved shortly after PEA, but the improvement of gas exchange was delayed for a period of time. Tanabe et al. ⁹ reported that the improvement of gas exchange capacity occurred 6–24 months after PEA surgery, and ‘steal blood’ gradually improved with the improvement of hypoxia. Therefore, we further speculate that this ‘steal blood’ phenomenon will gradually disappear over time due to the re-matching of ventilation/perfusion.

Study limitations

There are several limitations in this study. First, this study is a single center, retrospective and observational study. Although there are meaningful results at present, due to the small sample size of the included study, it is not excluded that the study results are biased to some extent. Therefore, multi-center and large-sample randomized controlled trials should be conducted in the future. Second, we used SPECT/CT to analyze the defect score at the level of segmental artery to reflect the therapeutic effect of BPA. However, BPA is usually performed across several segmental and sub-segmental areas, so perfusion defect analysis based on subsegmental level should be further explored in the future. And a more robust evaluation, namely, a quantitative absolute perfusion assessment, should be developed in order to avoid subjective interpretation biases. Third, most patients received long-term treatment with targeted drugs for pulmonary hypertension, which may affect our observation of the effect of BPA on changes in pulmonary perfusion.

Conclusion

This study shows that BPA not only improves the perfusion of target blood vessels, but also influences the perfusion of non-target blood vessels to a certain extent. This discovery may improve our understanding of the mechanism of BPA treatment, and help to develop new strategies for the treatment of CTEPH.