Chronic thromboembolic pulmonary hypertension (CTEPH) is a rare disease characterized by thrombotic and remodeling changes in pulmonary vessels resulting in a progressive, but potentially curable, form of pulmonary hypertension (PH). The condition affects 3.2-4.0% of patients who survive pulmonary embolism (PE),1–3 although it is diagnosed without a prior history of an acute event in 25.2% of cases.4 The pathophysiology of CTEPH involves occlusive organized thromboembolic material causing mechanical arterial obstruction as well as microvascular remodeling. The precise mechanisms leading to vascular remodeling are still unclear, but dysfunctional fibrinolysis, bronchopulmonary venous shunting, endothelial dysfunction, inflammatory mechanisms and genetic susceptibility appear to play a role.5,6 The molecular mechanisms of small-vessel disease partially overlap with those of pulmonary arterial hypertension (PAH).7 These changes ultimately lead to increased pulmonary vascular resistance (PVR), increased pulmonary arterial pressure and, eventually, right heart failure. The symptoms are mostly related to progressive right ventricular dysfunction, and are thus non-specific and initially present only on exertion, making early diagnosis challenging.8 As a result, a median delay of 14.1 months between initial symptoms and diagnosis has been reported4 and concerns about underdiagnosis have been raised.9 Diagnosis is obtained by precise measurement of pulmonary arterial pressures and evidence of vascular obstruction and/or perfusion defects.10

The treatment of choice for operable CTEPH is pulmonary endarterectomy (PEA). This is a complex surgical procedure with non-negligible perioperative mortality that frequently achieves normalization of hemodynamics and long-term symptomatic benefit.11,12 Despite being a potential cure for CTEPH, PEA is not suitable for all patients; only 63-68% of patients are technically eligible for PEA,4,13 as extensive involvement of the distal vessels may not be surgically accessible and discordance between PVR and surgically accessible obstructions may indicate extensive secondary vasculopathy that is not treatable by this approach.14 Underlying comorbidities and high surgical risk may also preclude the procedure, as well as patients’ personal preference, resulting in reported surgical rates of 50-58%.4,13,15

Two treatment modalities are available for inoperable CTEPH (as well as residual or recurrent PH after PEA): targeted therapy with pulmonary vasodilators, and balloon pulmonary angioplasty (BPA). Riociguat, a soluble guanylate cyclase stimulator, is the sole targeted therapy approved for CTEPH, improving hemodynamics and providing sustained benefits in exercise and functional capacity.16–18 However, off-label use of drugs approved for PAH may also be justified.10,19,20 BPA is an emerging treatment option in which small balloons are used to open obstructed vessels and widen stenotic lesions, over multiple sessions. It provides hemodynamic, functional and symptomatic benefits to inoperable patients.21,22

The available therapies for CTEPH are not mutually exclusive and can be used in combination,14,23 and the most suitable treatment should be decided in a multidisciplinary team setting.10,14

The aims of this study were to characterize a cohort of CTEPH patients followed in a referral center, to assess long-term outcomes according to the chosen treatment modality, and to identify predictors of mortality and morbidity.

CTEPH was mostly identified by CTPA or lung ventilation/perfusion SPECT. The latter presented better sensitivity (100%) in our cohort than CTPA (68.6%) and confirmed the diagnosis after a negative CTPA in 11 patients. Although 86.7% of patients underwent catheter-based pulmonary angiography to confirm the diagnosis and to assess operability, only one patient was diagnosed exclusively by this method.

Follow-up

During follow-up 11 patients died (24.4%), two (13.3%) in the PEA group and nine (30.0%) of those selected for pharmacological treatment. One-, two- and three-year survival was 93.3%, 82.4% and 75.9%, respectively. Kaplan-Meier curves for overall survival are presented in Figure 2. Although mortality was numerically higher in those proposed for pharmacological therapy, this did not achieve statistical significance (log-rank p=0.135; hazard ratio [HR] 0.327). However, after one perioperative death was excluded, PEA was associated with lower all-cause mortality (log-rank p=0.046; HR 0.158).

Figure 2.

Kaplan-Meier curves for the primary (A) and secondary endpoint (B) stratified by medical or surgical approach.

(0.11MB).

Regarding the secondary endpoint, the surgical approach was associated with a better prognosis, as shown in Figure 2 (log-rank p=0.036; HR 0.232). Two patients (13.3%) were hospitalized for decompensated heart failure or died, fewer than in the non-surgical group (n=13; 43.3%).

Serum BNP (HR 1.003; 95% confidence interval [CI] 1.001-1.005; p=0.003) and creatinine (HR 12.092; 95% CI: 1.121-130.390; p=0.040) at diagnosis were independently associated with the primary endpoint using a multivariate Cox regression model (Table 3). Regarding the secondary endpoint, PEA was associated with better outcomes (HR 0.198; 95% CI: 0.040-0.982; p=0.047), and BNP (HR 1.003; 95% CI: 1.001-1.005; p=0.008) and mPAP (HR 1.073; 95% CI: 1.022-1.128; p=0.005) at diagnosis were identified as independent predictors of the event.

Table 3.

Multivariate analysis for predictors of the primary and secondary endpoints.

	Variable	HR (95% CI)	p
Primary endpoint	BNP (pg/ml)	1.003 (1.001-1.005)	0.003
	Creatinine (mg/dl)	12.092 (1.121-130.390)	0.040
Secondary endpoint	BNP (pg/ml)	1.003 (1.001-1.005)	0.008
	mPAP (mmHg)	1.073 (1.022-1.128)	0.005
	PEA	0.198 (0.040-0.982)	0.047

The Cox proportional hazards model included as variables age, gender, history of acute pulmonary embolism, WHO functional class III/IV, BNP, serum creatinine, distance in six-minute walk test, mPAP, pulmonary vascular resistance, and pulmonary endarterectomy.

BNP: B-type natriuretic peptide; CI: confidence interval; HR: hazard ratio; mPAP: mean pulmonary arterial pressure; PEA: pulmonary endarterectomy.

Hemodynamics of operated patients improved: post-surgical mPAP in this group was 27.7±14.3 mmHg, with a statistically significant decrease of 23.8 mmHg (95% CI: 11.8-35.8; p=0.002). Residual pulmonary hypertension was identified in three of the patients who survived the procedure (21.4%), requiring pulmonary vasodilator therapy during follow-up. WHO FC also improved in 92.9% of patients, with most patients in class I following PEA (64.3%).

In this paper, our data illustrated the symptomatic burden of CTEPH, reinforced the positive impact of PEA on hemodynamics, mortality and morbidity, and identified characteristics associated with a worse prognosis.

Although low awareness of CTEPH among physicians could have led to later referral of patients to our center, the baseline characteristics of our patients, namely age, gender, functional class, hemodynamic profile and 6MWT, were similar to previous large studies on CTEPH, suggesting the same disease progression status as in other centers.4,11,25–27

We found that PEA provided better prognosis and a numerical trend for lower mortality. This is not surprising considering the poor prognosis for inoperable CTEPH, which has three-year survival of 70% as documented in the international CTEPH registry.25 In comparison, three-year survival for patients who underwent PEA was 89%.25 Recently, Kallonen et al.28 compared survival after PEA to survival in the general population at 15 years. The authors found that although survival was shorter in the PEA group, the difference in life expectancy between the groups was small.

When perioperative death was excluded, the survival rate of patients undergoing PEA was higher than that of pharmacological treatment. This finding is not unexpected as most mortality is known to be immediately associated with the complex procedure. Surgical experience is a necessary requirement for better outcomes, as shown by the results of Korsholm et al.,29 as in-hospital mortality decreased from 22.6% in the first decade to 4.3% in the second decade of experience, and from the United Kingdom National Cohort,12 which showed significant improvement in 30-day mortality from the first half of the cohort to the latest (13.2 to 2.4%, respectively).

Regarding overall prognosis, we identified baseline serum BNP and mPAP as well as inoperability as independent predictors of worse outcomes. This finding is in line with previously published studies.15,30,31 Other commonly identified risk factors for poor prognosis, like cardiac index and distance on 6MWT,13,30 did not show an independent association with worse prognosis in our study, probably due to the small sample size.

Surgical treatment was refused by 20% of our patients. This emphasizes the difficulty of treating CTEPH without a national surgical center and the reluctance of the population to go abroad for a complex operating, resulting in only one third of patients undergoing PEA. While this figure is still below the numbers reported in larger studies,4,25 it shows an improvement since 2013, when only 15.2% of incident CTEPH patients in a nationwide registry underwent the procedure.32 Such increases have also recently been reported by other PH treatment centers in the country, with 29.2-50% of patients undergoing PEA,33–35 probably reflecting better access to the procedure. Previously, PEA was not routinely available in Portugal and patients were referred to foreign centers, but the initial experience of a Portuguese PEA center was recently published.24

In the last two decades BPA has emerged as a novel alternative for inoperable CTEPH.36,37 In our cohort only two patients underwent this procedure after being considered inoperable by the PEA center's multidisciplinary team. Considering the very promising results from the Japanese and French registries, establishing benefits in symptom relief, hemodynamics and survival in inoperable patients,22,38 and the recent availability of the procedure in Portugal,39,40 increased referral for BPA can be expected in the future.

Our study presents some limitations. The small sample size from a single center is the most obvious, and partially limits our analysis of outcomes. Other limitations are the fact that patients underwent PEA at two different centers with different levels of expertise; the low rate of operated patients; and the differences in baseline characteristics between the two groups. The changes in care and referral of CTEPH patients over the last decade may also affect the interpretation of our results and may limit the external validity of our sample.

CTEPH is a condition with a poor prognosis but is potentially curable. PEA provided a better overall prognosis than non-surgical therapy, improving symptoms and hemodynamics, with a numerical trend for lower mortality. Our results reinforce the importance of (1) referral centers for disease diagnosis and management and (2) prompt referral to PEA centers of patients that fit operability criteria. Baseline hemodynamics and serum BNP at diagnosis were associated with higher mortality in CTEPH patients.

This study was financed by national funds via FCT (Foundation for Science and Technology) through the Strategic Project UID/NEU/04539/2019, UIDB/04539/2020, UIDP/04539/2020 (CIBB) and FEDER (RIGHT-2H POCI-01-0145-FEDER-032414).

The authors have no conflicts of interest to declare.