Results From the United States Chronic Thromboembolic Pulmonary Hypertension Registry

Background The United States Chronic Thromboembolic Pulmonary Hypertension Registry (US-CTEPH-R) was designed to characterize the demographic characteristics, evaluation, clinical course, and outcomes of surgical and nonsurgical therapies for patients with chronic thromboembolic pulmonary hypertension. Research Question What are the differences in baseline characteristics and 1-year outcomes between operated and nonoperated subjects? Study Design and Methods This study describes a multicenter, prospective, longitudinal, observational registry of patients newly diagnosed (< 6 months) with CTEPH. Inclusion criteria required a mean pulmonary artery pressure ≥ 25 mm Hg documented by right heart catheterization and radiologic confirmation of CTEPH. Between 2015 and 2018, a total of 750 patients were enrolled and followed up biannually until 2019. Results Most patients with CTEPH (87.9%) reported a history of acute pulmonary embolism. CTEPH diagnosis delays were frequent (median, 10 months), and most patients reported World Health Organization functional class 3 status at enrollment with a median mean pulmonary artery pressure of 44 mm Hg. The registry cohort was subdivided into Operable patients undergoing pulmonary thromboendarterectomy (PTE) surgery (n = 566), Operable patients who did not undergo surgery (n = 88), and those who were Inoperable (n = 96). Inoperable patients were older than Operated patients; less likely to be obese; have a DVT history, non-type O blood group, or thrombophilia; and more likely to have COPD or a history of cancer. PTE resulted in a median pulmonary vascular resistance decline from 6.9 to 2.6 Wood units (P < .001) with a 3.9% in-hospital mortality. At 1-year follow-up, Operated patients were less likely treated with oxygen, diuretics, or pulmonary hypertension-targeted therapy compared with Inoperable patients. A larger percentage of Operated patients were World Health Organization functional class 1 or 2 at 1 year (82.9%) compared with the Inoperable (48.2%) and Operable/No Surgery (56%) groups (P < .001). Interpretation Differences exist in the clinical characteristics between patients who exhibited operable CTEPH and those who were inoperable, with the most favorable 1-year outcomes in those who underwent PTE surgery. Clinical Trial Registration ClinicalTrials.gov; No.: NCT02429284; URL: www.clinicaltrials.gov.

Chronic thromboembolic pulmonary hypertension (CTEPH), an important diagnostic subgroup of pulmonary hypertension (PH), is characterized by the obstruction of pulmonary arteries with fibrotic, organized thrombus material and vascular remodeling resulting in PH and right ventricular failure. 1 Although CTEPH shares some clinical and pathologic characteristics with pulmonary arterial hypertension (PAH), the etiology, diagnosis, and treatment of CTEPH differ substantially from those of PAH.
Although many registries have focused on PAH, [2][3][4][5][6] there has not been a US multicenter registry focused on CTEPH. Investigators of the European CTEPH Registry described the epidemiology, risk factors, and outcomes of newly diagnosed European and Canadian patients with CTEPH. 7,8 However, the evolution in medical and surgical approaches, including the availability of riociguat and balloon pulmonary angioplasty (BPA), as well as the differences in CTEPH management between the United States and Europe, provided the rationale to organize the first US CTEPH registry.
The United States Chronic Thromboembolic Pulmonary Hypertension Registry (US-CTEPH-R) is a contemporary CTEPH registry involving 30 PH centers. The current report describes the demographic characteristics, medical history, symptoms, timeline to diagnosis, risk factors, diagnostic approach, disease management, and 1-year outcomes of 750 US patients newly diagnosed with CTEPH.

Patients and Methods
The US-CTEPH-R is a multicenter, prospective, longitudinal registry of patients newly diagnosed with CTEPH. The University of California San Diego (UCSD) is the sponsor and coordinating institution for the study, approved by the UCSD Human Research Protection Program (Project #141379). Thirty US sites (e-Appendix 1), selected based on a feasibility survey and geographic distribution, participated in the registry. Following provision of written informed consent, each patient was assigned a unique numerical patientidentifier to maintain confidentiality as required by the Health Insurance Portability and Accountability Act. The first patient was enrolled in April 2015, and target enrollment of 750 patients was met in March 2018. All subjects were followed up biannually until the last subject completed 1 year of follow-up in March 2019.
Consecutive patients diagnosed with CTEPH within 6 months of study consent and meeting the inclusion criteria were offered participation in the study. Time of diagnosis was defined as the date that the last of three hemodynamic and radiologic entry criteria for CTEPH (right heart catheterization, ventilation and perfusion lung scans, and pulmonary angiography) was met. Hemodynamic criteria included a mean pulmonary artery pressure (mPAP) $ 25 mm Hg measured by right heart catheterization and pulmonary capillary wedge pressure # 15 mm Hg, or > 15 mm Hg if justified by the investigator. Radiologic criteria included mismatched perfusion Take-home Points Study Question: What are the demographic characteristics, risk factors, diagnostic findings, treatment patterns, and 1-year outcomes of patients with CTEPH, and are there differences between operated and nonoperated patients? Results: US patients with CTEPH are predominantly middle-aged men and women with a history of PE. The diagnosis of CTEPH was often delayed because CTEPH symptoms were attributed to co-morbid conditions. Most patients in the US-CTEPH-R underwent PTE, with a 3.9% in-hospital mortality rate. PTE patients were younger, more likely of non-Hispanic White race, and less likely to have comorbidities than technically operable patients who did not undergo PTE. One-year outcomes were more favorable in Operated patients. Interpretation: The US-CTEPH-R provides a contemporary description of US patients with CTEPH, their evaluation, management, and 1-year outcomes. The registry also offers the opportunity for future investigation into the impact of comorbidities and medical therapies on long-term outcomes.
defects on ventilation and perfusion scanning and findings compatible with chronic thromboembolic disease on digital subtraction (DSA), CT angiography, or magnetic resonance angiography. A physician adjudication team (e-Appendix 2) reviewed imaging, with two adjudicators agreeing independently that imaging met the predefined criteria for CTEPH. Subjects meeting enrollment criteria were followed up longitudinally. Patients who had previously undergone pulmonary thromboendarterectomy (PTE) or BPA were excluded.
All evaluations and procedures performed were at the discretion of the treating clinician. Data collected at enrollment are given in the text and tables. Longitudinal data were collected biannually during patient clinic visits or by patient telephone call and/or chart abstraction.
Descriptive analysis provided the median and first/third quartiles for continuous variables and a frequency table reported for categorical variables. For comparisons among the three groups, a Kruskal-Wallis test was performed for continuous variables and Fisher exact test for categorical variables. The Wilcoxon rank sum test and Fisher exact test were used for between-group (pairwise) comparisons when the overall group comparisons differed significantly (P < .05). To account for multiple testing, Bonferroni's correction was applied, in which the critical P value for each pairwise between-group comparison was defined as P < .0167. Subgroup analysis of Operated and Operable/No Surgery patients comparing covariates of age, sex, race, and comorbidities identified by the Fisher exact test to be significantly different between the groups was performed by using univariate logistic regression analysis. R 3.5.2 statistical software (www. rproject.org; R Foundation for Statistical Computing) was used for analysis.

Results
Between April 2015 and March 2018, a total of 803 patients consented to participate in the US-CTEPH-R. Fifty subjects (6.2%) failed radiologic adjudication. The most common reasons included: acute rather than chronic pulmonary embolism (PE; n ¼ 18), no chronic PE (n ¼ 16), or advanced parenchymal lung disease (n ¼ 12). Three other subjects were excluded, resulting in 750 subjects in the final analysis (Fig 1) with subjects from all 50 US states (e-Appendix 3).
included subject refusing or delaying surgery (60%), risk exceeded benefit (19%) most commonly due to comorbidities, and patient too well (6.8%) with multiple reasons for not undergoing surgery, including advanced age listed for some. The remaining 566 patients (74.3% of the total cohort) underwent PTE at 20 US centers, 18 of which were US-CTEPH-R sites.

Demographic Characteristics
Median age at enrollment was 59 years (range, 16-87 years). Male subjects comprised 50.8% of the total cohort (Table 1). Inoperable and Operable/No Surgery subjects were older than Operated subjects (both, P < .001). The median BMI was 29.7 kg/m 2 (range, 12.3-58.7 kg/m 2 ), and Inoperable subjects had a lower BMI. Most subjects were WHO functional class III at enrollment (Fig 2). There was no significant difference in WHO functional class or 6-min walk distance (6MWD) among the three groups.

History of VTE Disease
The majority of patients with CTEPH (87.9%) reported a history of acute PE with (46.3%) or without (41.6%) DVT treated with anticoagulant agents. Few received systemic (2.3%) or catheter-directed (3.5%) thrombolytics. Almost one-half of the total CTEPH cohort reported recurrent episodes of PE (e- Table 2). The remaining patients either reported no history of acute PE or DVT (8.9%) or a history of DVT without PE (3.2%). Isolated upper extremity DVT was uncommon (1.6%). History of VTE in an immediate family member was reported in 17.7% of patients.
The median time from onset of cardiopulmonary symptoms until clinical suspicion of CTEPH was 10.0 months (interquartile range, 3.0-26.1 months) for the total study population. Although 70% of subjects reported persistent symptoms following treatment of  Tables 3 and 4). Obesity was more common in both Operable cohorts. COPD co-existed more commonly with CTEPH in the Inoperable cohort (24%) and the Operable/No Surgery cohort (23.9%) than in the Operated cohort (12.4%). patients with CTEPH who underwent PTE were also less likely to report a history of cancer.

Pulmonary Thromboendarterectomy
PTE was performed on 566 US-CTEPH-R patients. Proximal chronic thromboembolic disease (main, interlobar, or lobar pulmonary artery) was noted in most subjects, with involvement of the right lung (70%) more often than the left lung (54%). An  Figure 3 -Medical therapy at enrollment and 1-year follow-up. At enrollment, there is no difference in the percentage of patients in each cohort using oxygen or diuretics. More Inoperable patients are on PH-targeted therapy than Operated patients at enrollment (P < .001). At 1-year follow-up, significantly more Inoperable patients are on diuretics and/or oxygen than Operated patients (P <.001) and more Inoperable and Operable/No Surgery patients are being treated with PH-targeted therapy compared with operated patients (P < .001). PH ¼ pulmonary hypertension.

Baseline 1-Year Follow-Up
chestjournal.org additional surgical procedure was performed at the time of PTE in 184 subjects. Closure of a patent foramen ovale or atrial septal defect (n ¼ 87) was the most common secondary procedure, followed by coronary artery bypass graft (n ¼ 48) or tricuspid valve repair (n ¼ 20). Early postoperative measurements showed significant improvement in all hemodynamic parameters compared with baseline (e- Forty subjects (5.3%) underwent BPA, including 27 who were Inoperable, 6 who were operable but did not undergo PTE, and 7 who underwent both PTE and BPA.
Although WHO functional class distribution was similar in the three groups at baseline, at 1-year follow-up, the Operated group had a statistically better WHO functional class compared with the Inoperable and Operated/No Surgery cohorts (P < .001) (Fig 2). More of the Operated group were WHO functional class I or II at 1 year (82.9%) compared with the Inoperable (48.2%) or Operable/No Surgery (56%) cohorts (P < .001). However, all three cohorts showed functional class improvements at 1 year.
Fewer Operated subjects were on oxygen, diuretics, or PH-targeted therapy compared with the Inoperable group at 1 year (P < .001). The use of PH-targeted therapy decreased from baseline in the Operated group and increased at 1 year in the Inoperable and Operated/ No Surgery groups (Fig 3).
One-year mortality differed for the three cohorts (Operated, 5.6%; Operable/No Surgery, 9.9%; and Inoperable, 12.4%; P ¼ .028). Most deaths in the Operated group were perioperative, with three deaths following discharge attributed to thromboembolism. Six of the 10 deaths in the Inoperable group were attributed to worsening PH/right heart failure, and five of the eight deaths in the Operable/No Surgery group were attributed to worsening PH/right heart failure with one death due to hemorrhage.

Discussion
The US-CTEPH-R is the first prospective, multicenter, observational registry of US patients newly diagnosed with CTEPH. The US-CTEPH-R provides a contemporary profile of US patients with CTEPH, including epidemiologic and clinical features, results of diagnostic tests, medical therapy, characteristics of patients selected for PTE and those deemed Inoperable, and outcomes 1 year after enrollment.
An initial observation from the registry was that the diagnosis of CTEPH can be difficult. Despite study sites being experienced PH centers, 50 patients initially diagnosed with CTEPH (6.2%) failed adjudication, the differentiation of acute PE from chronic thromboembolic disease the most common reason for adjudication failure.
As has been reported previously, 7,10 diagnostic delays were common, with a median period of 10 months following symptom onset, and even a longer period of 13.1 months following an acute PE episode, prior to CTEPH consideration. Clinicians initially attributed the cardiopulmonary symptoms of almost one-half of patients to co-morbid diagnoses, such as obesity, COPD, or asthma. The potential consequence of doing so was that patients were often diagnosed at an advanced stage of disease. At enrollment, most subjects were WHO functional class III, and > 75% of subjects had a mPAP > 35 mm Hg with elevated pulmonary vascular resistance and reduced cardiac index. More than onehalf of the patients required diuretics, and almost onehalf were treated with PH-targeted therapy and/or supplemental oxygen.
Most registry patients reported a history of acute PE, with only 8.9% reporting no prior VTE. Patients commonly described incomplete resolution of Co-morbidities associated with CTEPH described in other cohort publications 9,11,12 were similarly observed in our registry, including a history of thrombophilia, cancer, non-type O blood group, and splenectomy. Possible reasons for the difference include the preenrollment adjudication of subjects in the US-CTEPH-R, differences in the definition of operability, and a large number of surgical centers participating in the US-CTEPH-R. For example, at UCSD, only 5.5% of the enrolled subjects were deemed Inoperable, reflecting a referral bias to a large surgical center.
Several study limitations have been identified. This analysis was an attempt at a "real-world" description of the evaluation of patients with CTEPH in the United States. There are differences between centers in what defines technical operability as well as co-morbidities that prohibit surgical candidacy. This resulted in some overlap between the Inoperable and Operable/No Surgery groups. US-CTEPH-R sites are academic PH centers, offering more advanced PAH and CTEPH therapies. Hence, the population enrolled by this registry likely does not reflect the CTEPH population cared for in the community without referral to a PH center. Many of the sites were also surgical centers, which may increase the percentage of Operable patients in the US-CTEPH-R. BPA was a newly emerging therapy in the United States during the conduct of the trial, and the small number of BPA subjects in the US-CTEPH-R does not likely reflect the role BPA currently plays in the management of CTEPH. Medical history was collected through chart abstraction and patient interview, with potential for recall errors. In addition, because US-CTEPH-R was observational, not all data elements were available for 1-year follow-up. One-year follow-up data are presented as descriptive statistics. Longer term follow-up data will be presented in a future publication and will include adjustments for covariates that may affect outcomes.

Interpretation
The US-CTEPH-R provides a contemporary description of US patients with CTEPH, their evaluation, management, and 1-year outcomes. Important observations include difficulties in making a timely diagnosis of CTEPH and the presence of co-morbidities that may mask the diagnosis and alter management. Nevertheless, US-CTEPH-R data show that PTE produces favorable outcomes with a low perioperative mortality for selected patients with CTEPH.