Periprocedural Complications With Balloon Pulmonary Angioplasty

Abstract

BACKGROUND

Balloon pulmonary angioplasty (BPA) was introduced as a treatment modality for patients with inoperable, medically refractory chronic thromboembolic pulmonary hypertension decades ago; however, reports of high rates of pulmonary vascular injury have led to considerable refinement in procedural technique.

OBJECTIVES

The authors sought to better understand the evolution of BPA procedure-related complications over time.

METHODS

The authors conducted a systematic review of original articles published by pulmonary hypertension centers globally and performed a pooled cohort analysis of procedure-related outcomes with BPA.

RESULTS

This systematic review identified 26 published articles from 18 countries worldwide from 2013 to 2022. A total of 1,714 patients underwent 7,561 total BPA procedures with an average follow up of 7.3 months. From the first period (2013–2017) to the second period (2018–2022), the cumulative incidence of hemoptysis/vascular injury decreased from 14.1% (474/3,351) to 7.7% (233/3,029) (P < 0.01); lung injury/reperfusion edema decreased from 11.3% (377/3,351) to 1.4% (57/3,943) (P < 0.01); invasive mechanical ventilation decreased from 0.7% (23/3,195) to 0.1% (4/3,062) (P < 0.01); and mortality decreased from 2.0% (13/636) to 0.8% (8/1,071) (P < 0.01).

CONCLUSIONS

Procedure-related complications with BPA, including hemoptysis/vascular injury, lung injury/reperfusion edema, mechanical ventilation, and death, were less common in the second period (2018–2022), compared with first period (2013–2017), likely from refinement in patient and lesion selection and procedural technique over time.

CENTRAL ILLUSTRATION

Balloon Pulmonary Angioplasty Complications Over Time

Procedure-related complications with balloon pulmonary angioplasty.

Balloon pulmonary angioplasty (BPA) continues to grow as a treatment option for patients with chronic thromboembolic pulmonary hypertension (CTEPH). Although BPA has demonstrated efficacy since its inception more than 4 decades ago, in 2001, Feinstein et al¹ reported a 6% mortality rate and a ~17% mechanical ventilation rate, thus halting the rapid adoption of BPA. A decade later in 2012, Matsubara and colleagues (Mizoguchi et al²) from Okayama Medical Center reported much more acceptable safety outcomes for BPA using a refined procedural technique.³ This report and others reignited BPA, and in 2018, updated guidelines from the Cologne Consensus Conference recommended a role for BPA in nonoperable, medically refractory CTEPH.^4,5 These guidelines marked a transition of BPA, which now has a Class 1 recommendation for inoperable and/or residual CTEPH.⁶

Pulmonary hypertension (PH) centers across the globe have published their individual outcomes with BPA; however, most reports have been limited to smaller populations and, in isolation, are unable to provide strong guidance on the global trends in procedural safety. To address this knowledge gap, we performed a systematic review and pooled analysis of all published articles reporting BPA outcomes worldwide since the Matsubara report² and compared procedure-related complication rates between the first 5-year period (2013–2017) and the second 5-year period (2018–2022). We performed a subgroup analysis excluding studies with an overlapping time frame between the 2 periods.

METHODS

This systematic review was created to assess safety outcomes of BPA for CTEPH patients in global PH centers. The statistical protocol was performed using prespecified subgroups of the first 5-year period (2013–2017) and the second 5-year period (2018–2022). The last year in which BPA was performed in each study was used for classification purposes. This systematic review is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses standards (PRISMA).⁷ The PRISMA checklist was followed.⁷ Ethics approval was obtained by each study.

SEARCH STRATEGY AND SELECTION CRITERIA.

Electronic databases (PubMed.gov, Scopus.com, EMBASE [including Embase Classic, via Embase.com], CINAHL Complete [EbscoHost], and Web of Science Core Collection [SCI-EXPANDED, SSCI, A&HCI, CPCI-S, CPCI-SSH, BKCI-S, BKCI-SSH, ESCI, CCR-EXPANDED]) were searched between January 1, 2000, to August 30, 2022, to include published reports of BPA. A systematic, sensitive search strategy was developed in collaboration with a research librarian (W.T.) using the following concepts: population (adult patients with inoperable CTEPH), concept (BPA complications), and context (change in complication rate over time). Detailed description of search strategy is provided (Supplemental Appendix).

SCREENING SOURCES OF EVIDENCE.

Two authors (M.A.S., D.B.) independently assessed the title and abstract of each study identified in the initial search to determine suitability for full-text review. Studies were selected if they included adults $18 years old receiving BPA for inoperable or residual CTEPH and reported primary outcome (incidence of at least 1 complication described in the following text). Publications were limited to those involving human subjects and in the English language. Duplicates, case reports, reviews, and abstracts were excluded. Any article deemed appropriate by either author was advanced to the next stage. Next, 4 authors (N.J., M.A.S., D.B., V.A.) independently reviewed the full text of all selected manuscripts to only include studies meeting prespecified criteria (Table 1). For this systematic review, the largest study from each center that included detailed BPA procedure-related complication rates were included in the final analysis to avoid redundant patient population data. Each full text was reviewed by 2 authors independently, and disagreements were resolved by a third author.

TABLE 1.

Study Inclusion and Exclusion Criteria

Inclusions	Exclusions
English language printed manuscript on BPA experience including clinical trials	Case reports or case series with <5 patients, abstracts, review articles
Largest study selected from each PH center to avoid redundant patient population	Incomplete data
Study outcomes were reported	Studies published before 2012

BPA = balloon pulmonary angioplasty; PH = pulmonary hypertension.

OUTCOMES.

The primary outcome was the incidence of BPA-related complications, including vessel injury or hemoptysis, use of invasive mechanical ventilation, and death based on known complications of the procedure.¹ Outcomes were assessed in prespecified cohort groups to compare overall rates in the first period (2013–2017) to the second period (2018–2022). Outcomes of vessel injury or hemoptysis, lung injury or reperfusion edema, and use of mechanical ventilation were reported as percentage per BPA session, whereas mortality was reported as a percentage per patient. A subgroup analysis was performed excluding studies with an overlapping time frame between the 2 cohorts.

STATISTICS.

The primary analysis was performed using t-test for each variable within a prespecified subgroup. Descriptive statistics were calculated using Stata IC (version 11.1, StataCorp).

RESULTS

A total of 4,825 studies were evaluated using titles and abstracts, with 164 undergoing full-text review (Figure 1). Overall, 26 studies that reported procedural and clinical outcomes on patients undergoing balloon pulmonary angioplasty met our study inclusion criteria and were included in this pooled analysis (Table 1). Our final analysis includes 26 reports from 4 continents and 18 countries, including 1,714 CTEPH patients undergoing a total of 7,561 BPA sessions. Average follow-up was 7.3 months after BPA per patient (Table 2).

FIGURE 1. Flow Diagram of the Search for Studies.

Flow diagram of the search for studies included in the systematic review according to the preferred reporting items for systematic reviews and meta- analysis statement. See Table 1 for inclusion and exclusion criteria.

TABLE 2.

Key Characteristics of All Studies Included in the Pooled Analysis

Location (First Author)	Study Years	N	BPA Sessions, n	Mean Follow-Up, mo
First period (2013–2017)
Multicenter Japan (Ogawa)20	2004–2013	308	1,408	14.2
Osaka, Japan (Ogo)21	2011–2015	80	385	12
Nagasaki, Japan (Yamagata)22a	2013–2016	11	31	1
Madrid, Spain (Velázquez)23	2013–2017	46	156	15.3
Hanover, Germany (Maschke)24	2013–2017	67	266	1
France (Brenot)8	2014–2017	184	1,006	6.1
Samsung, Korea (Kwon)25	2015–2017	15	52	6
Israel (Segel)26	2017	5	47	8
First period, total		716	3,351	7.9
Second period (2018–2022)
Leuven, Belgium (Godinas)27	2014–2018	18	91	1
Mayo, USA (Anand)28	2014–2018	31	75	NR
UK (Hoole)13	2015–2018	30	95	3
Hyderabad, India (Reddy)29	2017–2018	11	30	2
Multicenter Poland (Darocha)30	2013–2019	236	1,056	5.9
Bad Nauheim, Germany (Wiedenroth)31	2014–2019	142	780	6
Amsterdam, the Netherlands (van Thor)32	2015–2019	38	172	6
Paris, France (Jaiïs)9	2016–2019	52	400	12
Multicenter Japan (Kawakami)10	2016–2019	32	147	12
Chile (Sepúlveda)33	2016–2019	22	81	17.3
Almada, Portugal (Calé)34	2017–2019	11	57	6
Beijing, China (Jin)14	2018–2019	25	40	0
Grenoble, France (Piliero)17	2013–2020	156	646	6
Taiwan (Chen)35	2014–2020	13	59	21
Istanbul, Turkey (Atas)36	2017–2020	26	91	3
Guangzhou, China (Hong)37	2019–2021	92	210	0
San Diego, USA (Mahmud)38	2021	39	NR	NR
Athens, Greece (Karyofyllis)39	2015–2022	24	180	NR
Second period, total		998	4,210	6.7
Combined periods		1,714	7,561	7.3

^aOnly non-elderly cohort was included.

BPA = balloon pulmonary angioplasty; NR = not reported.

Pooled analysis showed a cumulative incidence of hemoptysis and/ or vascular injury of 11.1% (707/6,380), reperfusion edema rate of 6.0% (434/7,294), mechanical ventilation rate of 0.4% (27/6,257) per BPA session, and an overall 1.3% (21/1,707) mortality rate per patient (Table 3).

TABLE 3.

Pooled Complication Rates

Location (First Author)	Vessel Injury and Hemoptysis	Reperfusion Pulmonary Edema/Lung Injury	Mechanical Ventilation	Mortality
First period (2013–2017)
Multicenter Japan (Ogawa)20	244/1,408 (17.3)	251/1,408 (17.8)	17/1,408 (1.2)	12/308 (3.9)
Osaka, Japan (Ogo)21	47/385 (12.2)	18/385 (4.7)	2/385 (0.5)	NR
Nagasaki, Japan (Yamagata)22a	7/31 (22.6)	1/31 (3.2)	0/31 (0)	0/11 (0)
Madrid, Spain (Velázquez)23	34/156 (21.8)	9/156 (5.8)	NR	1/46 (1.0)
Hanover, Germany (Maschke)24	16/266 (6.0)	1/266 (0.4)	0/266 (0)	0/67 (0)
France (Brenot)8	118/1,006 (11.7)	92/1,006 (9.1)	4/1,006 (0.4)	0/184 (0)
Samsung, Korea (Kwon)25	4/52 (7.7)	2/52 (3.8)	0/52 (0)	0/15 (0)
Israel (Segel)26	4/47 (8.5)	3/47 (6.4)	0/47 (0)	0/5 (0)
Pooled (first period)	474/3,351 (14.1)	377/3,351 (11.3)	23/3,195 (0.7)	13/636 (2.0)
Second period (2018–2022)
Leuven, Belgium (Godinas)27	6/91 (6.6)	2/91 (2.2)	NR	0/91 (0)
Mayo, USA (Anand)28	3/75 (4.0)	1/75 (1.3)	1/75 (1.3)	1/31 (3.2)
UK (Hoole)13	NR	NR	NR	0/30 (0)
Hyderabad, India (Reddy)29	NR	2/30 (6.7)	0/30 (0)	0/11 (0)
Multicenter Poland (Darocha)30	NR	68/1,056 (6.4)	2/1,056 (0.19)	4/236 (1.7)
Bad Nauheim, Germany (Wiedenroth)31	84/780 (10.8)	115/780 (14.7)	1/780 (0.1)	2/142 (1.4)
Amsterdam, the Netherlands (van Thor)32	17/172 (9.9)	NR	NR	0/38 (0)
Paris, France (Jaïs)9	8/400 (2)	18/400 (4.5)	NR	0/52 (0)
Multicenter Japan (Kawakami)10	13/147 (8.8)	1/147 (0.7)	0/147 (0)	0/32 (0)
Chile (Sepúlveda)33	12/81 (14.8)	1/81 (1.2)	NR	0/22 (0)
Almada, Portugal (Calé)34	6/57 (10.5)	3/57 (5.3)	0/57 (0)	0/11 (0)
Beijing, China (Jin)14	0/40 (0)	0/40 (0)	NR	0/25 (0)
Grenoble, France (Piliero)17	23/646 (3.6)	19/646 (2.9)	0/646 (0)	1/156 (0.6)
Taiwan (Chen)35	6/59 (10.2)	3/59 (5.1)	NR	0/13 (0)
Istanbul, Turkey (Atas)36	14/91 (15.4)	3/91 (3.3)	0/91 (0)	0/26 (0)
Guangzhou, China (Hong)37	15/210 (7.1)	6/210 (2.9)	NR	0/92 (0)
San Diego, USA (Mahmud)38	NR (9.5)	NR	NR (0)	0/39 (0)
Athens, Greece (Karyofyllis)39	26/180 (14.4)	6/180 (3.3)	0/180 (0)	0/24 (0)
Pooled, second period)	233/3,029 (7.7)	57/3,943 (1.4)	4/3,062 (0.1)	8/1,071 (0.8)
Pooled, combined periods	707/6,380 (11.1)	434/7,294 (6.0)	27/6,257 (0.4)	21/1,707 (1.3)
P value comparison of periods	<0.01	<0.01	<0.01	<0.01

Values are n/total N (%) unless otherwise indicated.

^aOnly non-elderly cohort was included.

BPA = balloon pulmonary angioplasty; NR = not reported.

From the first period (2013–2017) to the second period (2018–2022), the cumulative incidence of hemoptysis and/ or vascular injury decreased from 14.1% (474/3,351) to 7.7% (233/3,029) (P < 0.01); reperfusion edema decreased from 11.3% (377/3,351) to 1.4% (57/3,943) (P < 0.01); invasive mechanical ventilation decreased from 0.7% (23/3,195) to 0.1% (4/3,062) (P < 0.01); and mortality decreased from 2.0% (13/636) to 0.8% (8/1,071) (P < 0.01) (Table 3, Central Illustration).

In the subgroup analysis excluding studies that overlapped between time periods, in the second period (2018–2022), compared with the first period (2013–2017), the cumulative incidence of hemoptysis and/or vascular injury decreased from 14.1% (474/3,351) to 6.0% (15/250) (P < 0.01); reperfusion edema incidence decreased from 11.3% (377/3,351) to 2.4% (6/250) (P < 0.01); mortality decreased from 2.0% (13/636) to 0% (0/156) (P < 0.01) (Table 4).

TABLE 4.

Subgroup Analysis Excluding Overlapping Studies

	Vessel Injury and Hemoptysis	Reperfusion Pulmonary Edema/Lung Injury	Mechanical Ventilation	Mortality	P Value
First period (2013–2017)	474/3,351 (14.1)	377/3,351 (11.3)	23/3,195 (0.7)	13/636 (2.0)	<0.01
Second period (2018–2022)	15/250 (6.0)	6/250 (2.4)	NR (0)	0/156 (0)	<0.01

Values are n/N (%). Unable to perform statistical analysis on mechanical ventilation.

DISCUSSION

Balloon pulmonary angioplasty is now an established treatment modality for inoperable and medically refractory CTEPH. Our pooled analysis suggests a decreasing incidence of procedure-related complications with BPA over time. More specifically, we observed a significant decline in hemoptysis and/or vascular injury, reperfusion edema, mechanical ventilation, and death rates with BPA between the first period (2013–2017) and the second period (2018–2022). These findings are reassuring and suggest a growing role of BPA in conjunction with pulmonary vasodilator therapy and anticoagulation, for patients with inoperable and medically refractory CTEPH.

We believe the decreased complication rates are largely attributable to refinement in technique and patient selection.^5,8 Our findings add to the literature that complications decrease with increased experience performing BPA. A group from Poland illustrated similar findings by describing the learning curve associated with their center’s experience with BPA, citing avoidance of “pouch” lesions, use of less aggressive anticoagulation strategies, and modification of catheter, guidewire, and lesion selection in order to reduce complication rates.⁵ Similarly, a group from France found over a 50% reduction in lung injury, and over an 80% reduction in severe lung injury, noting that patients who experienced lung injury had significantly higher baseline mean pulmonary artery pressure and pulmonary vascular resistance, and shorter 6-minute walking distance, implying that patient selection for BPA may need to be refined.⁸ Similar to the Polish cohort, lower complication rates were attributed to improved patient selection and refinement in BPA strategy by use of “knuckle wire” technique, undersized balloons on initial sessions, and use of a pressure wire–guided technique.⁸

Advances in BPA have been coupled with a paucity of treatment alternatives for inoperable CTEPH patients. In addition to lifelong anticoagulation, the primary alternative to BPA is a medication-only treatment approach with agents such a riociguat, the only Food and Drug Administration–approved pulmonary vasodilator medication for CTEPH.^5,9–11 Recently, new evidence from 2 clinical trials, which studied riociguat compared with BPA in a crossover study design, found that riociguat therapy before BPA may provide similar reductions in pulmonary vascular resistance compared with upfront BPA, as well as reducing subsequent procedure-associated complication rates.^4,9,10 Beyond hemodynamic and clinical parameters, select studies have also identified an improvement in quality of life and improved cardiopulmonary exercise parameters following BPA.^12–14 Additionally, Japanese and German centers found improvements in right ventricular function assessed via cardiac magnetic resonance findings BPA, although these studies were limited by small sample size.^15,16

STUDY LIMITATIONS.

Limitations of our analysis include those intrinsic to all retrospective systematic reviews and pooled cohort analyses, including publication bias, as PH centers with high complication rates may be less likely to publish complication outcomes. Furthermore, many studies have reported increased complications during their early experience with BPA which may have skewed their complication rates⁵; however, for the purposes of this pooled analysis, we were unable to account for such variations. The lack of standardized accepted definitions for each outcome across included studies is another major limitation of this analysis. This is particularly true for the outcome of reperfusion pulmonary edema, whose definition has evolved over time and this evolution is likely contributing to a reduction in reported rates for this outcome in more recently published reports. Furthermore, hemoptysis severity was not a reported outcome in included studies because granular data on the more severe forms of hemoptysis such as those requiring intensive care admission were not available. In regard to generalizability, centers that perform more BPA may have more experienced providers and equipment, which may skew data toward lower rates of complications that may not be generalizable to all PH centers.¹⁷ Some of these centers have different techniques, including routine use of intravenous ultrasound and pressure wires, which some studies suggest may result in lower complication rates, although more research is needed to confirm these findings.^18,19 Finally, there are studies from the 2 cohorts that overlapped in time, particularly those near the 2018 cutoff year but also beyond this time frame. For this reason, a subgroup analysis was performed (Table 4) that excluded studies that overlapped between the 2 time periods. This analysis still yielded a statistically significant difference between cohorts (P < 0.01).

CONCLUSIONS

BPA is undergoing natural evolution and becoming safer with increasing adoption over the years. Future work will further refine procedural technique and patient and lesion selection for BPA.

PERSPECTIVES.

WHAT IS KNOWN?

BPA is now an established treatment modality for patients with inoperable, medically refractory chronic thromboembolic pulmonary hypertension, although early studies report high procedural complication rates.

WHAT IS NEW?

This study demonstrates that periprocedural complications with BPA have significantly reduced over time.

WHAT IS NEXT?

Future work will continue to further refine patient selection and procedural technique with BPA as well as advise uniform metrics and standardized definitions for this emerging field.

ABBREVIATIONS AND ACRONYMS

BPA

balloon pulmonary angioplasty

CTEPH

chronic thromboembolic pulmonary hypertension

PH

pulmonary hypertension

APPENDIX

For an expanded Methods section, please see the online version of this paper.