Article Text


Original article
Balloon pulmonary angioplasty in patients with inoperable chronic thromboembolic pulmonary hypertension
  1. Arne K Andreassen1,
  2. Asgrimur Ragnarsson1,
  3. Einar Gude1,
  4. Odd Geiran2,
  5. Rune Andersen3
  1. 1Department of Cardiology, Oslo University Hospital Rikshospitalet, Oslo, Norway
  2. 2Department of Cardiovascular and Thoracic Surgery, Oslo University Hospital Rikshospitalet, Oslo, Norway
  3. 3Department of Radiology, Oslo University Hospital Rikshospitalet, Oslo, Norway
  1. Correspondence to Arne K Andreassen, Department of Cardiology, Oslo University Hospital Rikshospitalet, Oslo N-0027, Norway; aandreas{at}


Objective To examine the effect of balloon pulmonary angioplasty (BPA) on chronic thromboembolic pulmonary hypertension (CTEPH) in patients with inoperable disease or persistent pulmonary hypertension after pulmonary endarterectomy.

Design Observational cohort study.

Setting Referred patients with inoperable or persistent CTEPH.

Patients Twenty consecutive CTEPH patients (10 females), aged 60±10 years.

Interventions BPA.

Main outcome measures Right heart catheterisation, functional capacity (cardiopulmonary exercise testing (CPET) and NYHA class) and blood sampled biomarkers N-terminal pro-brain natriuretic peptide (NT-proBNP) and troponin T examined at the time of diagnosis and repeated in all patients 3 months after the last BPA.

Results Seventy-three catheterisations were performed with 18.6±6.1 BPAs per patient on segmental and subsegmental arteries. Two deaths occurred following the first BPA, with an overall 10% periprocedural death rate. Reperfusion oedema complicated seven procedures. Comparisons before and after BPA showed significant haemodynamic improvements, including decreased mean pulmonary artery pressure (mPAP) (45±11 mm Hg vs 33±10 mm Hg; p<0.001) and increased cardiac output (4.9±1.6 L/min vs 5.4±1.9 L/min; p=0.011). Reduced right ventricular strain was indicated by significantly lower plasma levels of NT-proBNP and troponin T. Significant improvement in functional capacity was evident as assessed by NYHA class (3.0±0.5 vs 2.0±0.5; p<0.001) and CPET (13.6±5.6 mL/kg/min vs 17.0±6.5 mL/kg/min; p<0.001). Seventeen patients (85%) were alive after 51±30 months of follow-up.

Conclusions BPA may offer an alternative form of treatment in selected CTEPH patients. While prognostic markers such as haemodynamics, functional capacity and biomarkers improve, significant periprocedural complications must be recognised. Randomised trials are warranted.

Statistics from


Chronic thromboembolic pulmonary hypertension (CTEPH) is characterised by obstruction of the pulmonary vascular bed by organised thromboembolic material, with subsequent vascular remodelling in smaller vessels.1 Increased vascular resistance with progressive vascular pulmonary hypertension leads to right heart strain and failure. While our understanding of the pathophysiological mechanisms remains incomplete, the prognosis of CTEPH is poor and related to the pulmonary artery pressure (PAP) and the degree of right ventricular failure.

Pulmonary endarterectomy is considered the treatment of choice in CTEPH and has curative potential.2 However, this surgical procedure may not be offered to patients with either substantial distal vascular obstruction or significant comorbidity. Recent data from an international prospective registry demonstrated that more than a third of patients were considered inoperable.3 Parallel with the introduction of new drugs and improved care in pulmonary arterial hypertension, inoperable CTEPH patients are increasingly being offered this same medication. However, conflicting results concerning haemodynamics and functional capacity have been published and cautious use of medical therapy is endorsed in current guidelines.4

Balloon pulmonary angioplasty (BPA) has been suggested as an alternative invasive procedure in patients not amenable to surgery.5–7 Despite reported improvements in both haemodynamics and functional capacity, BPA is still considered an experimental treatment.1 We conducted an investigation of BPA in CTEPH patients either deemed inoperable or with sustained postoperative pulmonary hypertension, to determine if we could reproduce and expand on the experiences of a former series of patients who underwent this procedure.6 Repeat right heart catheterisation, biomarkers and cardiopulmonary exercise testing (CPET) were performed in all patients, who served as their own controls.


This study conforms to the STrengthening the Reporting of OBservational studies in Epidemiology recommendations for reporting cohort studies.8

Study patients

The diagnosis of CTEPH was confirmed by the radiological detection of vascular obstructions, followed by right heart catheterisation. While a ventilation/perfusion lung scan had been performed in most local hospitals, surgical suitability was determined by both pulmonary CT angiography and conventional invasive pulmonary angiography in all patients at our centre. Selective coronary angiography was undertaken in all but two patients and all were assessed with echocardiography. Possible candidates for this prospective trial between January 2003 and August 2011 were referred symptomatic CTEPH patients 18–75 years of age. Fifty patients underwent pulmonary endarterectomy during this period. The location of the thromboembolic lesions, comorbidities and age were considered by a qualified team of thoracic surgeons and cardiologists when deciding if the condition was operable or not. Distal pulmonary artery obstructions considered surgically inaccessible were found in 16 patients. Three patients rejected surgery, while two were considered poor candidates for pulmonary endarterectomy because of high age, diabetes and systemic hypertension. The youngest participant had undergone prior surgery with persistent pulmonary hypertension. Patients gave informed consent before the first intervention. The limited experience with and experimental nature of BPA was emphasised to candidates, and the study was conducted in line with the Declaration of Helsinki and approved by the local ethics committee.

Haemodynamic studies

Right heart catheterisation was performed according to standard procedures. Right atrial, pulmonary artery and pulmonary capillary wedge pressure (PCWP), and cardiac output (CO) (in triplicate) were determined using a flow-directed thermodilution catheter advanced into the pulmonary artery through a jugular or femoral vein. All participants had a mean PAP (mPAP)>25 mm Hg and PCWP ≤ 12 mm Hg. A catheter inserted into the right radial artery was used for systolic blood pressure and arterial blood gas monitoring. Pulmonary vascular resistance (PVR), cardiac index (CI) and stroke volume (SV) were calculated according to standard formulas.

Cardiopulmonary exercise test

A maximal, graded, exercise test was performed on an upright electrical braked bicycle ergometer (Jarger ER900; Viasys Healthcare, Hochberg, Germany). The initial power was 20 W and work load was ramped up at 10 W/min. Patients were encouraged to exercise as long as possible and instructed to maintain ∼60 rpm, but could stop at any time. Oxygen consumption (VO2), CO2 production (VCO2) and ventilation (ventilatory efficiency, VE) were measured on a breath-to-breath basis (MVmax 229; Viasys Healthcare). Heart rate and ECG were continuously recorded. Blood pressure was measured at the end of each step. For parameters sampled continuously or breath-by-breath, average values for 20 s time intervals were used in the analysis. Peak VO2 and respiratory exchange ratio (RER) were measured in the last 20 s interval during exercise. The anaerobic threshold was defined as the VO2 at the point where ventilation started to increase exponentially compared to the increase in VO2. The oxygen pulse was calculated by dividing VO2 by heart rate. Ventilatory efficiency on exercise was shown by the VE/VCO2 slope, determined by linear regression analysis of all data obtained in the VE/VCO2 plot during exercise. MVmax 229 software was used for calculation.

Measurement of biomarkers

Morning blood samples for the measurement of N-terminal pro-brain natriuretic peptide (NT-proBNP) and troponin T were drawn from a peripheral vein. NT-proBNP was determined by an electrochemiluminescence immunoassay on a modular platform (Roche Diagnostics, Basel, Switzerland). Plasma troponin T levels were assessed with an extremely sensitive quantitative test, with a detection limit for troponin T of 0.01 µg/L, and concentrations ≥0.01 µg/L considered abnormal, with such patients termed troponin T positive.

Balloon pulmonary angioplasty

Pulmonary artery catheterisations were performed by two senior interventional radiologists (AR and RA) with extensive experience in coronary and general vessel interventions. A standard procedure was typically performed from the right femoral vein, with anticoagulation continued but with a reduced dosage of warfarin to maintain an INR∼2.0. A Radiofocus guide wire M 0.035″ with an angled tip (Terumo, Tokyo, Japan) was advanced ahead of either a standard pigtail angiography catheter or a Judkins right coronary guiding catheter 0.071″ (JR 5, Launcher; Medtronic, Minneapolis, Minnesota, USA). Selective and super-selective catheterisation of lobar arteries, segmental and when appropriate subsegmental arteries was performed to demonstrate stenosis, occlusions and reduction of parenchymal opacification after injection of vascular contrast medium Omnipaque 140 mgI/mL (GE Healthcare, Oslo, Norway) or Iomeron 300 mgI/mL (Bracco, Milan, Italy). Following administration of intravenous heparin 5000 IE (supplied with 2500 IE every 60 minutes), stenotic lesions in the form of total occlusions, filling defects or intravascular bands were passed with 0.014″ micro guidewires (Balance Middle Weight Universal II; Abbott, Santa Clara, California, USA). Coronary monorail balloon catheters (diameter 1.25–4.50 mm) and renal monorail balloon catheters (diameter 5.0–7.0 mm) were positioned over the selected lesion and inflated with a mixture of saline and contrast medium to pressures of 1–10 atmospheres for 5–30 s. Variations in inflation pressure were chosen according to balloon behaviour during inflation and vessel size. Immediate selective angiography was performed after dilations to confirm satisfactory results and rule out residual stenosis or signs of vessel damage. The guide wire remained in its distal position during contrast injection. Dilations were repeated with unchanged or increased balloon size in case of lesions not responding with a near normalisation of angiographic vessel size. A typical catheterisation with BPAs lasted for 1.5–2 h. In order to reduce the possibility of reperfusion oedema, each treatment session was confined to three lung segments. Following BPA procedures, patients were moved to our ICU for 2–4 h surveillance, before telemetric observation on the normal ward for up to 48 h. Lesions of other lung segments deemed accessible for further BPA treatment were assessed after 6–8 weeks.

Statistical analysis

Analysis was performed with SPSS statistical software V.18.0 (SPSS, Chicago, Illinois, USA) and a two-sided p value<0.05 was considered statistically significant. Paired changes in baseline parameters to 3 months after their last BPA were calculated for all patients. The Student t test was used for all normally distributed variables and the Mann–Whitney test for other continuous variables. Results are reported as mean±SD or as otherwise appropriate.


Patients and BPA activity

Twenty consecutively recruited patients underwent BPA (table 1). Anticoagulation with warfarin had been initiated at the local hospital. Two patients were on targeted therapy with sildenafil that was terminated before their first BPA, and none received any specific medication for pulmonary arterial hypertension at follow-up evaluation. Thrombotic risk factors were present in four patients: two had lupus anticoagulant antibodies, one had polycytemia vera and one had protein C and S deficiency.

Table 1

Patients’ characteristics at diagnosis

Seventy-three catheterisations were performed, averaging 3.7±2.1 procedures per patient (range: 2–9). Of the total number of 371 BPAs, 118 were performed on segmental arteries and 253 on subsegmental arteries, with an average of 18.6±6.1 per patient (figure 1). Seven cases of reperfusion pulmonary oedema were treated with supplementary oxygen and diuretics. One patient died within 2 h after his first BPA and right ventricular failure was verified post mortem. Another died 9 days after his first BPA with clinical signs of acute pulmonary embolism. The mean follow-up of the 18 other patients was 51±30 months. At 3 months following their last BPA, NYHA functional class improved, with the number of patients in classes I, II, III, and IV changing from 0/3/14/3 to 4/11/3/0, respectively (p<0.001). Two of the patients remaining in class III were considered for bilateral lung transplantation. This was successfully performed in one patient 16 months after his last BPA, but denied due to high age in the other, who died 15 months after his last BPA.

Figure 1

Pulmonary angiograms of right middle lobe subsegmental (A and B) and lower lobe segmental branches (C and D) before and immediately after balloon pulmonary angioplasty.

Haemodynamics and biomarkers

Haemodynamic and biomarker data are summarised in table 2. Compared to values at diagnosis, mPAP had decreased significantly in patients 3 months after their last BPA. In addition, CO increased significantly with normalisation, as did calculated CI and SV. PVR decreased by a mean of 33%, while systolic blood pressure and PCWP remained unchanged. In line with haemodynamic and respiratory improvements, we also noted a significant decline in resting heart rate. Saturations in both mixed venous and arterial blood increased significantly, and three of five patients on long term oxygen treatment were able to discontinue treatment after BPA. Plasma levels of NT-proBNP and troponin T fell significantly over time, with normalisation of NT-proBNP in eight patients and troponin T levels below the detection limits in our laboratory in all but three patients. The BPA effects on mPAP, SV, PVR, functional capacity and NT-proBNP are given for each patient in figure 2.

Figure 2

Individual effects of balloon pulmonary angioplasty on mean pulmonary artery pressure, stroke volume, pulmonary vascular resistance, functional capacity and N-terminal pro-brain natriuretic peptide. NT-proBNP, N-terminal pro-brain natriuretic peptide; PVR, pulmonary vascular resistance; SvO2, mixed venous oxygen saturation; VO2, oxygen consumption.

Table 2

Haemodynamics, blood gases and biomarkers at diagnosis and 3 months after the last BPA

Exercise capacity

Exercise performance improved significantly after BPA as evaluated by several CPET parameters (table 3). Thus, peak VO2, exercise duration, peak heart rate, peak O2 pulse and RER increased, while the VE/VCO2 slope declined at follow-up. Changes in maximal systolic blood pressure were not significant.

Table 3

Cardiopulmonary exercise testing at diagnosis and 3 months after the last BPA


We have demonstrated that BPA treatment in selected CTEPH patients: (i) may decrease PAP and improve haemodynamic profile; (ii) tends to normalise levels of biomarkers used to assess right ventricular function; (iii) enhances functional capacity; and (iv) is currently restricted to highly motivated patients because of its experimental nature and significant complications. Thus, we reproduce and extend the results of a former series of CTEPH patients treated with BPA.6 We believe that this less invasive procedure should be considered as a supplementary treatment option in CTEPH patients. Thus, while all CTEPH patients should be considered for pulmonary endarterectomy, suitable candidates for BPA are patients with inoperable distal disease, otherwise operable patients with comorbidities, and surgically treated patients with significant postoperative pulmonary hypertension.

CTEPH is a devastating disease, with pulmonary endarterectomy offering a clear survival benefit.9 Riedel et al10 followed 76 non-surgically treated CTEPH patients for up to 15 years. A presenting mPAP>30 mm Hg invariably led to progressive pulmonary hypertension, with a less than 20% survival at 2 years with mPAP>50 mm Hg. Similarly, in 49 patients only on anticoagulation therapy, a 3-year death rate of 90% was observed in those with mPAP>30 mm Hg.11 On the other hand, in the ASPIRE registry, 58 patients with inaccessible disease had a 5-year survival rate of about 60%.12 Also, among 48 Japanese CTEPH patients with mPAP=50 mm Hg, a mean survival time of 6–8 years was noted.13 Thus, even though data are somewhat conflicting, prognosis seems poor for those deemed inoperable or with persistent postoperative pulmonary hypertension related to deteriorating right ventricular function.10–13 As a consequence, pulmonary arterial hypertension-targeted therapy has become more common among CTEPH patients with inoperable disease.3 Small and uncontrolled trials give some support for medical intervention,14 but only a modest haemodynamic improvement with no gain in exercise capacity was demonstrated with bosentan in the only placebo-controlled study performed to date.15 Also in the AIR study, a controlled clinical trial that included patients with various forms of precapillary pulmonary hypertension, inhaled iloprost failed to show beneficial effects in the CTEPH subgroup.16 Therefore, no specific pharmacotherapy is currently approved for CTEPH.

Even though BPA may be viewed as a less complex intervention than pulmonary endarterectomy, the two periprocedural deaths we experienced underline its potentially serious side effects. As with novel medical treatments in general, implementing BPA in CTEPH patients has resulted in certain changes in strategies over time. First, supportive equipment during catheterisation and BPA has been altered and refined. Second, postprocedural surveillance is undertaken in all patients in the cardiac care unit, irrespective of the catheterisation laboratory experience. Except for the fatal case, reperfusion oedema was handled with supplementary oxygen and diuretics, without the need for intratracheal intubation or percutaneous cardiopulmonary support. Third, in order to prevent more extensive oedema secondary to an immediate rise in parenchymal perfusion, we believe that it is better to carry out multiple shorter procedures with a limited number of dilations. Interestingly, only one fatality was observed when intravascular ultrasound was included in the BPA procedures in a recent report on 68 patients with pre-BPA haemodynamics comparable with those in our group, suggesting improved safety with this device.17 Fourth, the development of pulmonary oedema following BPA appears to be linked to haemodynamics and right ventricular function, as all but one of our patients were in the highest quartile for mPAP, PVR and NT-proBNP and in the lowest quartile for SV and peak VO2. Thus, we hope these data will be useful for those caring for patients and help to significantly lower the current 10% periprocedural death rate.

With older patients with comparable baseline right heart catheterisation data, we observed more pronounced haemodynamic improvement compared to the 18 patients reported by Feinstein and colleagues.6 Our more extensive BPA activity, with more than three times as many balloon dilations performed per patient, may have led to better pulmonary blood flow distribution and greater reductions in right ventricular afterload. Thus, in contrast to the former study, we were able to show significant improvements in CO and PVR and higher oxygen saturations in arterial and mixed venous blood, in addition to our common reductions in pulmonary pressures. Interestingly, a significant fall in resting heart rate was observed after BPA, a haemodynamic variable associated with increased mortality in patients with established coronary artery disease and left ventricular dysfunction.18 Resting HR has also shown a similar prognostic value in idiopathic pulmonary arterial hypertension, and was recently included in a risk score calculator for pulmonary arterial hypertension.19 ,20 The use of biomarkers assessing ventricular wall stress (NT-proBNP) and myocardial injury (troponin T) support our haemodynamic data indicating a reduction in right ventricular strain, with markedly lower plasma levels after BPA. Both have been shown to be independent predictors of survival in precapillary forms of pulmonary hypertension.21 ,22

The haemodynamic changes seen following multiple BPAs relieved patients of dyspnoea and enabled them to improve their functional capacity, as assessed by CPET. In a national registry of 469 CTEPH patients, exercise capacity was an independent predictor of survival in the inoperable group, together with CI.23 Thus, improvement in both variables after BPA probably contributed to our observed transplantation-free survival rate of 80% after a mean follow-up of more than 4 years. In contrast to the other patients, the two who developed late signs of progressive right heart failure did not improve their exercise capacity, biomarkers or haemodynamics.

Matsuda et al also used CPET to evaluate functional capacity after endarterectomy.24 The peak oxygen consumption and pulmonary pressures in their patients were comparable to those in our older group before intervention, with a higher PVR in their group. The improvements among our patients in peak oxygen consumption and VE (VE/VCO2 slope) were similar to those seen in patients treated surgically and tested 1 month postoperatively. Whether we could have matched their 1-year data showing further improvement in exercise capacity, remains unanswered. However, we were unable to normalise pulmonary pressures in most patients and mildly elevated pressures might contribute to more symptomatic disease over time. We speculate that higher systemic/pulmonary blood flow during exercise after BPA treatment explains the decrease in VE/VCO2 slope and increase in peak O2 pulse previously described in idiopathic pulmonary arterial hypertension.25 Finally, an imbalance between sympathetic and parasympathetic activity affecting the heart rate profile has long been recognised to result in adverse cardiovascular events and death.26 Thus, the higher chronotropic reserve indicated by a lower pulse at rest and higher maximal heart rate during exercise after BPA treatment, may be protective and reflect an improvement in the underlying autonomic dysfunction in CTEPH patients.

The present study is subject to the limitations inherent to a non-randomised study with relatively few patients. While randomised trials are called for, it is not obvious what form of treatment BPA should be compared with. Because of the present uncertainties around pharmacotherapy, comparison with medical treatment is problematic but would be justified in drugs approved in the future. Riociguat is a stimulator of soluble guanylate cyclase and might be such a candidate, as a preliminary report showed promising effects on haemodynamics and functional capacity in inoperable CTEPH.27 As regards the use of surgery versus BPA, CTEPH patients with distal type III disease might be a suitable population for randomisation,2 but views on these issues will probably vary and complicate the development of proper protocols.

In conclusion, the long-term haemodynamic and functional improvements seen after BPA justify consideration of this invasive technique as a treatment option in CTEPH patients with surgically inaccessible disease. Keeping in mind their poor prognosis and current lack of therapeutic alternatives, we would strongly recommend experienced centres with dedicated colleagues skilled in invasive procedures to familiarise themselves with this approach. Randomised trials, although perhaps difficult to perform, are warranted.


View Abstract


  • Contributors AKA, AR and RA: conception and design, and analysis and interpretation of the data; AKA, EG and RA: drafting of the article: AKA, AR, EG, OG and RA: provision of study materials or patients, collection and assembly of data, and final approval of the article.

  • Competing interests None.

  • Ethics approval The local ethics committee approved this study.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.