Balloon Pulmonary Angioplasty in Chronic Totally Occluded Pulmonary Arteries: Applying Lessons Learned From the Treatment of Coronary Artery Chronic Total Occlusions

Objectives. This study aims to describe the safety and efficacy of revascularizing chronic total occlusions (CTOs) of the pulmonary arteries with balloon pulmonary angioplasty (BPA) in patients with chronic thromboembolic pulmonary hypertension (CTEPH). Background. BPA has emerged as an effective treatment for CTEPH patients when surgical treatment is not possible. Experience to date has suggested treating CTOs may be associated with excess risk and less procedural success relative to other lesion types. Methods. This study is a retrospective case series of all BPAs on CTOs for individuals with CTEPH at a single institution. Procedural approach, complications, and success rate over a 6-month period are described. Results. During the study period, 6 individuals with 15 CTOs were identified and intervened upon during 21 interventions. Success rate for revascularization was 62% per attempt and 87% per lesion. Techniques used for successful intervention include true to true lumen wiring (n = 7) and subintimal dissection re-entry with subintimal tracking and re-entry (n = 3), Stingray balloon (Boston Scientific) assisted re-entry (n = 2), and direct wire re-entry (n = 1). Wire perforations were relatively common and occurred in 62% of interventions, but rarely resulted in a change in clinical status. Conclusions. Although important barriers to routine intervention on CTOs in CTEPH remain, the current series suggests a higher success rate than previously reported experiences using CTO revascularization techniques including subintimal tracking and re-entry and Stingray balloon-assisted re-entry. Although the frequency of wire perforation was relatively high, the clinical ramifications of these complications were mild.

Key words: chronic total occlusions, chronic thromboembolic pulmonary hypertension

Over the past several years, balloon pulmonary angioplasty (BPA) for the treatment of chronic thromboembolic pulmonary hypertension (CTEPH) has emerged as an important alternative to pulmonary thromboendarterectomy (PTE) in patients with distal disease or high operative risk. Initial outcomes of BPA in this patient population, as reported by Feinstein et al, were suboptimal, with a high rate of morbidity and mortality.¹ It is only following more recent successes at a few pioneering centers that BPA has re-emerged as a promising therapeutic option. Undoubtedly, the checkered history of BPA for the treatment of CTEPH has resulted in a very careful and measured approach with regard to treatment strategy. It is now commonplace to undersize balloons and limit the number of segments treated per intervention so as to reduce the likelihood of significant pulmonary vascular or reperfusion injury. Similarly, the use of polymer-jacketed or high penetration force wires is often avoided.

As our understanding of and experience with BPA continues to grow, it remains imperative to ensure not only a safe procedure, but an effective one. Chronic total occlusion (CTO) of the pulmonary arteries represents the most severe occlusive pathology and the recommendation remains to proceed with PTE in patients with CTO whenever possible; however, there are many patients with significant impairment and CTOs who may not be operative candidates.² CTOs in particular present several challenges to percutaneous revascularization and caution has been recommended in addressing these lesions percutaneously. Lack of distal vessel visualization, surrounding alveoli, and tortuous or ambiguous anatomy create significant hurdles for safe transcatheter revascularization of these lesions. Despite these challenges, revascularizing chronically occluded segments offers the possibility of significant symptomatic and hemodynamic improvement through substantial vessel recruitment. It is with these benefits in mind that our group has devoted focused effort to tackling the challenges of CTO revascularization in this patient population. We now present our initial experience with targeted revascularization of chronically occluded pulmonary arteries for the treatment of CTEPH.

Study population. All CTEPH patients at the University of Washington in Seattle, Washington are reviewed at a monthly multidisciplinary conference that typically includes representatives from Thoracic Surgery, Pulmonary Vascular Medicine, Interventional Cardiology, and Radiology (Interventional Radiology, Nuclear Medicine, and/or Chest Radiology). The primary mandate of the group is to determine whether a patient is an operative candidate for PTE. If a patient is not considered to be an operative candidate, the group then decides whether BPA can be offered. All CTEPH patients who underwent BPA at our institution between August 2019 and February 2020, who were discovered to have a CTO and underwent attempted CTO revascularization, are included in this case series. We retrospectively report procedural success, procedural complications, and the consequences of these complications. The University of Washington institutional review board reviewed and approved this study (study #00010025). No financial support was received to conduct this study or for manuscript preparation.

Definitions. CTO lesions are defined as lesions with no contrast flow beyond a proximal occlusive cap during selective angiography. Procedural success is defined as angiographic patency of the occluded vessel with clearly identifiable pulmonary venous return. Procedural complications include any vascular injury resulting from the intervention, regardless of its impact on the clinical status of the patient. Mild hemoptysis is defined as hemosputum. Moderate hemoptysis is defined as <50 mL of blood loss as measured by expectorated blood and not including blood in the alveoli. The clinical impact of each procedural complication is reported, including the presence of hemoptysis, hypoxia, hemodynamic compromise, need for additional urgent procedures, and/or death.

Procedural approach. All procedures were performed in a cardiac catheterization laboratory under moderate sedation. All but 1 patient received heparin boluses during the procedure, with a goal activated clotting time of ≥200 seconds. One patient received intraprocedural anticoagulation with bivalirudin due to a history of heparin-induced thrombocytopenia.

All procedures were performed via femoral venous access with the following set-up. A 9 Fr, 10 cm Pinnacle sheath (Terumo) was inserted into the femoral vein. A 7 Fr balloon wedge catheter was advanced into the branch pulmonary arteries for pressure measurements and oximetry. The balloon wedge catheter was then exchanged for a 6 Fr, 90 cm Brite Tip sheath (Cordis), which was prolapsed through the 9 Fr short sheath and advanced into the branch pulmonary artery (PA) of interest. A 6 Fr, 110 cm Judkins Right 4 guide catheter (Boston Scientific) was inserted into the sheath and directed toward the occluded vessel. A 6 Fr Trapliner guide-extension catheter (Teleflex) and 0.018˝ microcatheters were used for additional wire support and to aid in visualization beyond the occlusion cap via microcatheter angiography. In the majority of cases, a 135 cm Corsair microcatheter (Asahi Intecc) was used. However, a 135 cm Turnpike Spiral microcatheter (Teleflex) and a 2.0 x 15 mm Apex over-the-wire compliant balloon (Boston Scientific) were used in select cases.

An antegrade true lumen to true lumen approach was attempted in all cases. We adhered to a wire-escalation technique utilized in the revascularization of chronically occluded coronary arteries³ with sequential upgrade of 0.014˝ coronary wires from spring coil to jacketed hydrophilic to high penetration power wires. Following wire penetration of the CTO cap, a microcatheter was advanced beyond the proximal occlusion and angiography through the microcatheter was performed. If an intraluminal position was confirmed, balloon angioplasty of the lesion was performed. Balloon dilations were performed using a maximum balloon diameter of 2.0 mm on vessels visually estimated between 2-3 mm in diameter, 4.0 mm on vessels between 4-5 mm in diameter, and 6.0 mm on vessels ≥6 mm in diameter. Balloon lengths ranged between 15-20 mm; however, serial dilations along the length of longer lesions were often required. No stents were implanted during any intervention. If angiography revealed an extravascular location, such as within the alveoli, bronchi, or pleural space (Figure 1), a wire was readvanced beyond the vascular perforation and the microcatheter was withdrawn. Balloon tamponade and/or heparin reversal were performed according to clinical need. If the microcatheter resided within the subintimal space, luminal re-entry was attempted.

At the conclusion of each procedure, all patients were transferred to the medical intensive care unit for overnight observation. Patients received bridging anticoagulation overnight, beginning 6 hours post hemostasis. Patients receiving vitamin K antagonists resumed their use on the day of the procedure with outpatient enoxaparin bridging the following day. Patients receiving direct oral anticoagulants resumed their use the morning following the procedure.

Statistical analysis. Descriptive summary statistics are presented, including mean and range as appropriate. Mean PA (mPA) pressure was calculated as 1/3 of the systolic PA (PAs) pressure and 2/3 of the diastolic PA (PAd) pressure. Pulmonary vascular resistance (PVR) was calculated as (mPA pressure minus the occlusion pressure)/cardiac output. Pulmonary vascular compliance (PVC) was calculated as the stroke volume/(PAs pressure – PAd pressure). Analyses were performed using STATA 15.1 (StataCorp).

Between August 1, 2019 and February 29, 2020, a total of 15 CTOs were identified in 6 patients who underwent 21 separate attempts at revascularization. Our cohort included a wide range of disease severity, with an average mPA pressure of 39 mm Hg (range, 21-53 mm Hg), cardiac index of 2.3 L/min/m² (range, 1.8 – 2.7 L/min/m²), and PVR of 5.9 WU (range, 2.5-10.7 WU). All 6 patients were on at least 1 pulmonary antihypertensive agent and 3 patients were on dual agents (Table 1). Most patients (4/6) had undergone previous PTE; however, none had undergone PTE within the previous year.

Of the 15 identified CTOs, 8 were located at the segmental level, 5 subsegmental, and 2 suprasegmental. Procedural success was achieved in 13 out of 21 interventions. Ten out of 13 lesions with a successful intervention required only 1 intervention to establish vessel patency. One lesion required 2 separate interventions and 2 lesions required 3 separate interventions in order to achieve success. Two lesions were not successfully revascularized on any attempt. Revascularization was attempted once for 1 of these lesions and twice in the other lesion. Of the 13 successful interventions, 7 lesions were wired from true lumen to true lumen using a Pilot 200 (Abbott Cardiovascular) (n = 4), Astato 20 (Asahi Intecc) (n = 2), or Sion guidewire (Asahi Intecc) (n = 1). In the remaining 6 lesions, the guidewire was recognized to be subintimal well beyond the occlusive cap. Re-establishment of intraluminal wire position was achieved in 3 interventions using subintimal tracking and re-entry (STAR) technique⁴ with Pilot 200 (n = 2) and Suoh wires (Asahi Intecc) (n = 1) wires. Luminal re-entry was achieved via antegrade dissection and re-entry (ADR) technique with a Stingray balloon (Boston Scientific) in 2 cases (Figure 2) and with direct wire re-entry using a Mongo wire (Asahi Intecc) in 1 case. Follow-up angiography was performed on 7 of the revascularized CTO lesions at a subsequent intervention at least 1 week post successful intervention. All 7 reimaged lesions were found to be widely patent. Follow-up angiography during subsequent sessions has not yet been performed on the remaining 6 lesions.

Eight interventions were unsuccessful at establishing vessel patency. In all 8 cases, CTO cap penetration was successful, but the subintima was entered. ADR technique with a Stingray balloon was attempted in 1 case, but failed due to an expanding subintimal layer as a result of microcatheter angiography. This lesion was successfully reintervened upon a week later via a repeated attempt of ADR with a Stingray balloon, after allowing for subintimal healing.

Procedural complications were relatively common, occurring in 13 out of 21 interventions (Table 2). The vast majority were due to vessel perforation into either the alveoli, bronchi, pulmonary veins, or pleura. One patient experienced a retrograde PA dissection during 2 separate interventions on the same lesion. Angiography following each intervention demonstrated complete resolution of each dissection without apparent sequelae. Of the 12 cases of vessel perforation, 8 required no treatment whatsoever and had no impact on case duration. In 1 of these 8 cases, the patient experienced mild hemoptysis, which spontaneously resolved. Four cases of perforation required active management. Two of these perforations resulted in moderate hemoptysis. Of these, 1 patient received heparin reversal and balloon tamponade, successfully sealing the perforation; however, the procedure was terminated early, and the patient spent 1 extra day in the hospital for observation. The patient made a full recovery and went on to receive additional BPA. The remainder of the 3 perforations were treated with some combination of balloon tamponade and/or heparin reversal. Following hemostasis, heparin was rebolused and all 3 procedures resumed without further incident. There were no procedural complications in 7 interventions. Mild hemoptysis in the absence of angiographic evidence of vascular injury occurred in 2 of these cases, both of which resolved spontaneously.

We present an initial experience for treating CTOs at a single institution. This experience suggests a relatively high rate of vessel perforation; however, the clinical ramifications of these complications were mild and the success rate in restoring vessel patency was relatively high.

Over the past decade, BPA treatment for CTEPH has shown great promise. Following early success by Japanese operators, BPA has become recognized as an important therapeutic intervention for patients with limited surgical options. Multiple retrospective studies on the efficacy of BPA have demonstrated sustained improvements in mPA pressures, PVR, cardiac output, right ventricular function, functional class, and peak oxygen consumption.^5-9 Despite this, there is limited published literature devoted to addressing the technical hurdles of treating complex PA occlusions such as CTOs. As BPA is increasingly regarded as an established treatment for CTEPH, achieving complete revascularization in individuals with more complex lesions is an important next step.

The relative paucity of publications on CTO interventions may reflect the technical challenges and perceived risk of revascularizing chronically occluded PAs. In their landmark 2016 publication, Kawakami et al reported on their experience treating 1936 lesions.¹⁰ Of these, only 67 (3.5%) were classified as CTOs, of which only 52% were successfully revascularized. While this may suggest a low incidence of CTO lesions, this study only reported the percentage of treated lesions that were CTOs. Thus, this report may more accurately reflect a hesitance to pursue BPA on complete occlusions.

Hesitation to proceed with interventions on CTOs is not without precedence. Taniguchi and colleagues had less success and more complications when intervening on CTOs relative to ring-like stenoses or webs.⁹ Similarly, Kurznyna and colleagues observed improved overall outcomes with a suite of changes, including avoidance of CTOs.¹¹ They also raised the question of whether intervention on CTOs should be avoided.

In contrast, our data suggest that CTO revascularization in patients with CTEPH is both safe and effective. Over a 6-month period, our group attempted revascularization on every CTO identified in an effort to understand and overcome the technical challenges posed by these lesions. Of the 21 attempts on 15 lesions, 13 attempts were successful, leading to a success rate of 62% per attempt and 87% per lesion. These results are not dissimilar to an abstract presented by Sethi and colleagues at the Transcatheter Cardiovascular Therapeutics (TCT) conference in 2018 or Roik and colleagues, who each report an 81% success rate in small case series of either CTOs or a mix of total and subtotal occlusions.¹²

Revascularization was most often accomplished with true lumen to true lumen wiring (7 out of 13) using either polymer-jacketed or high penetration force wires. However, entrance into the subintimal space was a relatively common occurrence with this technique, resulting in change in strategy to dissection re-entry. In 1 case, success was achieved via direct re-entry with a high penetration force wire. In the remaining 5 cases, we employed STAR and Stingray balloon-assisted re-entry. To our knowledge, this is the first report of successful PA-CTO revascularization using these 2 techniques.

While our approach to PA revascularization adhered to well-recognized techniques used in the treatment of coronary artery CTOs, several differences are recognized. First, the distal segments of occluded coronary arteries are often visualized via a network of collaterals. Occluded PAs rarely collateralize in this manner, and determining the location of a wire beyond the occlusion often cannot be accomplished with a retrograde angiogram. Our approach to this problem was to perform microcatheter angiography beyond the occlusive cap to identify whether an intraluminal position was achieved. Second, the complexity and three-dimensional nature of PA anatomy makes it very difficult to predict an occluded vessel’s course, increasing the likelihood of wire and microcatheter perforation. We approached this challenge by exchanging high penetration force wires for softer spring coil or jacketed wires with more predictable vessel tracking. Third, lack of distal vessel visualization raises questions as to whether the distal branches are filled with embolism, in situ thrombus, or remain patent beyond the proximal obstruction. With increasing experience, we recognized that the vast majority of unsuccessful attempts at true lumen to true lumen revascularizations were due to entry into the subintimal space (14 out of 21) rather than extensive thrombus distal to the occlusion cap. Thus, most of our ADR attempts were successful just distal to the occlusion cap, preserving the majority of subsegmental branches.

In our cohort, none of the occluded vessels received collateral flow. Therefore, microcatheter angiography served as the primary method of determining wire position beyond the occlusive cap, ensuring no balloon angioplasty would be performed within a perforated vessel. As expected, wire perforation was a relatively common occurrence; however, the clinical impact of these complications was minimal, similar to the experience reported by Inami and colleagues.¹³ The majority of perforations spontaneously resolved. Those that did not responded appropriately to anticoagulation reversal and balloon tamponade. While microcatheter perforation through an occluded coronary artery may lead to significant hemorrhage, we did not observe this in the PAs. The reason for this is not fully known. It may be an artifice of our still relatively small volume or may be due to foundational differences in the pathology of chronically occluded PAs. Although quite speculative, diseased pulmonary vasculature from chronic thromboemboli may have a very different elasticity compared with calcified coronary atherosclerosis, leading to significant lesion recoil and hemostasis following microcatheter removal.

One clear challenge for CTO intervention is interpretation of distal lesion pathology. For example, early interventions occasionally resulted in an angiographic pattern of contrast distal to the lesion within a vascular space, but without distal flow into the capillary bed or pulmonary veins. We initially interpreted this as heavy thrombotic disease beyond the proximal occlusion; however, attempts at distal balloon angioplasty never resulted in restoration of forward flow into the periphery or pulmonary venous return when this initial pattern was seen. It was only after repeated interventions that we recognized this as the angiographic pattern of subintimal dissection within the PAs (Figure 3). As our understanding of subintimal vessel anatomy evolved and our success rate improved, we found that re-entry into the PA lumen just beyond the occlusive cap with either STAR or Stingray balloon technique resulted in significant vessel recruitment. In hindsight, the vast majority of our failed interventions were due to misinterpretation of subintimal dissection as heavy thrombotic burden, offering the possibility for higher success rates during the initial intervention as we improve our ability to work within this space.

Study limitations. Our study has several limitations. First and foremost, our small numbers limit generalization as to the efficacy and complication rate of CTO intervention within the PAs. Furthermore, we are unable to accurately comment upon hemodynamic improvements following CTO interventions as the majority of these lesions were treated concomitantly with non-CTO lesions, obscuring interpretation of postprocedural hemodynamic data. One might expect greater improvements in cardiac output, PA pressures, and PVR following restoration of flow within total occlusions as opposed to partial stenoses; however, this has yet to be demonstrated and requires further study. Finally, when weighing the benefits of CTO intervention, catheterization lab exposures such as contrast use, radiation dose, and total procedure time must be evaluated in addition to success and complication rate. We are unable to accurately report these exposures due to concomitant interventions on non-CTO lesions.

Despite the limitations of this study, we have demonstrated that PA-CTO revascularization is possible using antegrade revascularization techniques pioneered in the treatment of coronary artery CTOs, resulting in a relatively high overall success rate. A learning curve clearly exists with complex lesion intervention and a number of the CTOs required repeated attempts at revascularization. With increasing experience, we believe CTO revascularizations may be performed with greater efficiency and may offer significant hemodynamic and symptomatic benefits to patients. Our experience suggests that, alongside refinements in approach and pattern recognition, BPA interventions on CTOs may be an increasingly viable way to restore blood flow in the pulmonary vasculature in individuals with CTEPH.

Transcatheter revascularization of CTOs in patients with CTEPH is technically possible, with a relatively high degree of success and low periprocedural morbidity. Subintimal entry is common but ADR with Stingray balloon and STAR techniques are effective strategies for luminal re-entry and recruitment of large territories of pulmonary blood flow.