Repeat pulmonary thromboendarterectomy outcomes: A 15-year single-center retrospective review

Objective: Chronic thromboembolic pulmonary hypertension is potentially curable via pulmonary thromboendarterectomy. A minority of patients experience recurrence of their symptoms and are eligible for repeat pulmonary thromboendarter-ectomy. However, little data exist regarding risk factors and outcomes for this patient population. Methods: We performed a retrospective review of the University of California San Diego chronic thromboembolic pulmonary hypertension quality improvement database, including all patients who underwent pulmonary thromboendarterec-tomy from December 2005 to December 2020. Of the 2019 cases performed during this period, 46 were repeat pulmonary thromboendarterectomy procedures. Demographics, preoperative and postoperative hemodynamics, and surgical complications were compared between the repeat pulmonary thromboendarterec-tomy group and 1008 ﬁ rst pulmonary thromboendarterectomy group. Results: Repeat pulmonary thromboendarterectomy recipients were more likely to be younger, to have an identi ﬁ ed hypercoagulable state, and to have higher preoperative right atrial pressure. Etiologies of recurrent disease include incomplete initial endarterectomy, discontinuation of anticoagulation (noncompliance or for medical reasons), and anticoagulation treatment failure. Patients who received repeat pulmonary thromboendarterectomy had signi ﬁ cant hemodynamic improvement, but less pronounced compared with patients who received ﬁ rst pulmonary thromboendarterectomy. Repeat pulmonary thromboendarterectomy was associated with an increased risk of postoperative bleeding

In patients with recurrent CTEPH, repeat PTE carries an increased risk of complications but results in significant hemodynamic improvement with comparable surgical mortality in an experienced center.

PERSPECTIVE
This study demonstrates that repeat PTE may be an appropriate, safe, and effective option in patients with CTEPH recurrence or incomplete initial PTE when performed at an experienced center.In addition, these findings may assist in risk stratification and facilitate improved joint decision making with patients regarding the risks and benefits of undergoing repeat PTE.
Chronic thromboembolic pulmonary hypertension (CTEPH) is a potentially curable form of pulmonary hypertension.CTEPH prevalence has been estimated as high as 38.4 per million inhabitants. 1 Prospective epidemiological data suggest that CTEPH incidence after acute, symptomatic pulmonary embolism can reach 4% with up to 2500 new cases of CTEPH diagnosed in the United States each year. 2,35][6][7][8] Despite surgical intervention, there have been reports of patients presenting with recurrent symptoms and new or residual surgically accessible disease.The etiology and risk factors of disease recurrence have not previously been described, and the optimal treatment course for this patient population is unknown.0][11] On the basis of these data, practitioners may defer referral for repeat PTE and focus on nonsurgical management.With the emergence of alternate treatment regimens, efficacious medical therapy such as riociguat, and expanding practice of balloon angioplasty (BPA), it is important to elucidate the true safety and efficacy of repeat PTE.This study was designed to describe the cause of recurrent thromboembolic disease in this patient population and to determine perioperative outcomes with repeat PTE surgery.

MATERIALS AND METHODS
We performed a retrospective review of the University of California San Diego (UCSD) CTEPH quality improvement database, including all patients (n ¼ 2019) who underwent PTE between December 2005 and December 2020.Of the 2019 procedures reviewed, 111 were identified as reentry sternotomy procedures.Of those, 46 cases were identified as repeat PTE.Of note, 2 of the identified cases were for a third PTE; thus, 44 unique patients had repeat PTE at UCSD with 46 total repeat PTE procedures.Results from second and third PTE procedures were recorded as separate cases.The 65 reentry sternotomy cases not due to PTE were excluded from final analysis.Of the 2019 procedures, 157 were missing complete hemodynamics; hemodynamic data in these patients were excluded from final analysis, but other recorded variables including demographics and postoperative complications were included.Final analysis included 1908 first PTE cases and 46 repeat PTE cases (Figure 1).
Preoperative hemodynamics were obtained in both first PTE and repeat PTE groups before the procedure.Postoperative hemodynamics were obtained in the intensive care unit (ICU) before removal of the pulmonary artery catheter.Pulmonary vascular resistance (PVR) was estimated substituting the right atrial pressure for the pulmonary artery wedge pressure.Proximal versus distal disease was determined by operative report with Jamieson types and later standardized UCSD Classification Levels as previously described. 12Proximal disease was defined as disease originating at the main, descending, or lobar pulmonary artery (level 1 or 2) in either the left or right lung.Postoperative bleeding was defined as any early mediastinal bleeding requiring transfusion of blood products or return to operating room.Reperfusion injury was defined as evidence of new pulmonary opacity in an endarterectomized region and hypoxemia with no other explanation for the hypoxemia or opacity.
All hemodynamic variables are reported as mean AE standard deviation.Two-sided Fisher exact test was performed for all categorical variables, and 2-tailed unpaired t tests with P value confirmation via the false discovery rate approach were performed for all continuous variables.Nonparametric Mann-Whitney testing was used to report ventilator days and ICU length of stay.Software used for statistics was GraphPad Prism version 8.00 for Windows (GraphPad Software, www.graphpad.com).This study, conducted using an existing quality improvement database, has been granted an exemption from the UCSD Institutional Review Board.

RESULTS
Patients undergoing repeat PTE were younger and more likely to have a history of acute pulmonary embolism and an identified hypercoagulable state than those undergoing first PTE.The most common hypercoagulable state identified in patients undergoing repeat PTE was antiphospholipid syndrome, as determined by the presence of anti-cardiolipin, anti-beta-2 glycoprotein1 antibodies, or lupus anticoagulant (n ¼ 19, 41.3%).In addition, patients receiving repeat PTE were more likely to be on preoperative pulmonary hypertension-targeted therapy (65.2% vs 48.2%, P ¼ .023).Preoperative right atrial pressures were higher in the repeat PTE group.Otherwise, preoperative hemodynamics did not statistically differ between the 2 groups (Table 1).The median time between first and second PTEs was 6 years (3; 9) (range, 1-28 years), and 46% (21/46) of patients had their first PTE at UCSD (Table E1).
Twelve patients (26%), all of whom had their initial procedures at other centers, underwent repeat PTE to correct residual surgically accessible disease and symptomatic pulmonary hypertension, likely due to incomplete initial resection.Recurrent thrombosis was the indication for PTE in the remaining 34 patients.The most common cause of recurrent thrombosis was treatment failure (warfarin ¼ 9, direct oral anticoagulant ¼ 4, and low-molecular-weight heparin ¼ 1).Anticoagulation noncompliance was documented in 11 patients (24%).Recurrent thrombosis occurred when anticoagulation was withheld for a medical indication: intracranial hemorrhage (n ¼ 2), pericardial effusion (n ¼ 1), hemoptysis (n ¼ 1), or other procedures (n ¼ 2; dental and orthopedic surgery).Three cases did not have a clear cause of recurrent thrombosis (Figure 2).Postoperative hemodynamics were substantially improved in both cohorts (Figure E1).However, when compared with the first PTE group, patients receiving repeat PTE had significantly higher postoperative mean pulmonary artery pressures, higher postoperative PVR, and total pulmonary resistance.Cardiopulmonary bypass time was significantly longer in the repeat PTE group compared with the first PTE group.Circulatory arrest time was longer in the repeat PTE group, but not statistically significant (Table 2 and Figure E2).Proximal disease was found in a similar percentage of first PTE and repeat PTE patients, and the majority (75%) of those with an initial incomplete PTE were found to have proximal disease at the time of repeat PTE surgery.
Patients receiving repeat PTE had higher median ventilator, ICU, and postoperative hospital days than the first PTE group.Patients receiving repeat PTE were more likely to have postoperative bleeding complications, reperfusion lung injury, and residual pulmonary hypertension as defined by postoperative PVR greater than 400 dyn s/cm -5 compared with the first PTE group.Frequency returning to the operating room was not statistically different between the 2 groups.Patients requiring repeat PTE were more likely to be discharged with pulmonary hypertensiontargeted therapy (Table 3).There was 1 in-hospital mortality (2.2%) among the 46 repeat PTE surgeries, which was not statistically significant compared with first PTE surgeries (1.9%, P ¼ .88).This subject died of refractory airway hemorrhage.

DISCUSSION
This is the largest reported series of repeat PTE.We found that repeat PTE may result in significant hemodynamic improvement and has comparable mortality when compared with those undergoing first PTE, but is associated with increased complications.In the appropriate patient population, repeat PTE is feasible and safe when performed at an experienced center.
PTE is a technically challenging operation requiring cardiopulmonary bypass with profound hypothermia and periods of circulatory arrest for complete bilateral pulmonary endarterectomy.In repeat PTE surgery, the procedure is even more challenging not only due to considerations of redo-sternotomy in the setting of severe pulmonary hypertension and right heart hypertrophy but also due to specific difficulties with establishing a plane of dissection within the pulmonary arteries.A team of multidisciplinary experts carefully review the history and course of the recurrent disease (Figure 3).Typically, whenever possible, we would like to wait more than 6 months from the first operation before performing a repeat PTE.

THOR
As part of the assessment for repeat PTE, the adequacy of the initial PTE is reviewed, including evaluation of the specimen retrieved, the degree of reperfusion after the initial surgery, and initial hemodynamics.Those with unfavorable outcomes after initial surgery with adequate specimen retrieval would not likely benefit from repeat PTE.The characteristics of the residual/recurrent disease are also evaluated, including the surgical proximity and degree of clot, and whether current clot burden can appropriately explain their pulmonary hemodynamics.Those with a small clot burden but severe pulmonary hypertension are not likely to benefit from repeat intervention.There are no absolute contraindications to repeat surgery, although relative contraindications may include a poorly tolerated initial PTE, prior history of postoperative thrombosis despite anticoagulation, or significant postoperative lung injury.In addition to the usual preoperative evaluation and workup, review of an updated computed tomography scan of the chest to identify mediastinal landmarks and anticipated challenges of redo sternotomy in the setting of right ventricular hypertension and hypertrophy is crucial.With careful preoperative planning and meticulous redo-sternotomy techniques, chances of cardiac injury are low; notably, no patients had cardiac injury intraoperatively in this cohort.All patients have intraoperative femoral arterial and venous access lines placed before sternotomy for possible immediate need of percutaneous initiation of cardiopulmonary bypass, although this is rarely needed.Once sternotomy is completed, the goal is to identify and clear the adhesions over the ascending aorta and the right atrium to initiate full bypass and decompress the heart.Once this is performed, further take-down of adhesions can be performed during the cooling phase.The procedure then continues in a similar fashion to a first-time PTE, although cardiopulmonary bypass time may be longer in this patient population.The endarterectomy portion of the procedure can be more complex in repeat PTE.In patients with inadequate or incomplete first operation, the endarterectomy plane is undisturbed and full PTE can be performed without any further challenge.However, in patients with prior complete endarterectomy and true recurrent disease, the new plane of dissection is typically deeper, more scarred, and more challenging to remove.It is not unusual to have to enter the plane of media so that endarterectomy can be performed.The challenges of identifying a new plane and completing full endarterectomy in these patients explain the longer cardiopulmonary bypass and circulatory arrest times.
More than a quarter of patients who underwent repeat PTE underwent first PTE at another center and, on evaluation at our center, were noted to have incomplete initial PTE.Notably, 75% of these patients were found to have proximal CTEPH at the time of their repeat procedure.This reaffirms the current recommendation that operability assessment and surgery should be performed at an

1516
The Journal of Thoracic and Cardiovascular Surgery c December 2023 Thoracic: Lung Astashchanka et al THOR experienced CTEPH center. 4,5Surgical candidacy and operability assessment remain subjective based on experience and controversial, differing from center to center.Those with residual pulmonary hypertension after PTE may benefit from evaluation at an expert center to determine repeat procedure candidacy.Anticoagulation noncompliance and anticoagulation failure account for the majority of recurrent CTEPH in the evaluated cohort.Lifelong anticoagulation is recommended after PTE for all patients, and vitamin K antagonists traditionally have been the recommended antithrombotic for both individuals with CTEPH and those with antiphospholipid syndrome. 12,13The risks and benefits of vitamin K antagonists versus direct-acting oral anticoagulants (DOACs) in the CTEPH population are limited to retrospective case series at this time.One evaluation of 501 patients with CTEPH did not demonstrate any significant difference in efficacy of DOACs versus warfarin with respect to recurrent thromboembolism, 14 whereas a retrospective review of 1000 patients post-PTE demonstrated a disproportionately higher incidence of thromboembolism recurrence in those using DOACs versus warfarin (4.62%/person-year for DOACs vs 0.76%/person-year for vitamin K antagonists, P ¼ .008). 14,15In addition, a recent abstract reports that patients with CTEPH referred for PTE on DOAC therapy were twice as likely to have associated acute or subacute thrombi intraoperatively. 16Additional research is needed to determine the safety and efficacy of DOACs in this patient population, but may be an important factor in enhancing patient compliance and preventing thrombosis reoccurrence.
8][19] In our cohort, residual pulmonary hypertension as we defined by a PVR greater than 400 dynes.sec.cmÀ5 on early postoperative hemodynamics was significantly more common in the repeat PTE population (35.0%) than in the first PTE population (11.8%).In addition, although our study also found significant hemodynamic improvement, patients receiving repeat PTE were more likely to be discharged from the hospital on pulmonary hypertensiontargeted therapy and oxygen than the first PTE group.Our findings are consistent with previous publications by Mo and colleagues 11 and Merli and colleagues, 10 who also found patients undergoing repeat PTE had less pronounced improvement in their mean pulmonary artery pressure, PVR, and total pulmonary resistance than those undergoing first time PTE.Despite this, the prior series demonstrated repeat PTE resulted in statistically significant improvement in their degree of pulmonary hypertension and symptoms as assessed by New York Heart Association functional class. 105][26] Many patients post-PTE who remain symptomatic may not be eligible for a repeat PTE due to the distal nature of the residual chronic thromboembolic disease.At our institution, approximately 20% to 25% of our BPA cohort are post-PTE; additionally, of our patients post-PTE BPA, 94% had their PTE performed at UCSD. 24 However, if there was an incomplete endarterectomy or recurrent thrombosis leading to new proximal occlusions, these patients may benefit from redo PTE instead because BPA may be less effective for proximal lesions and complete occlusions.
As demonstrated in prior studies evaluating repeat PTE, those undergoing repeat surgical intervention were more likely to have postoperative complications.However, unlike previous studies, repeat PTE in our current study was not associated with increased hospital mortality.In contrast to the original cohort study by Mo and colleagues, 11 reperfusion lung injury was more common in our cohort of repeat PTE cases compared with first-time PTE cases.Reperfusion lung injury typically occurs within 48 hours of revascularization and presents with hypoxemia and radiographic infiltrates in areas reperfused with surgery. 18Severity of preoperative pulmonary hypertension and the presence of residual pulmonary hypertension have been associated with an increased risk of reperfusion lung injury. 20,27Postoperative complications, including an increase in postoperative bleeding and reperfusion injury, are likely the source of the overall increase in the median ventilator days, ICU length of stay, and postoperative length of stay.Although mortality does not appear to be affected, the increased morbidity can add a significant cost burden to the patient.

Study Limitations
This study has several limitations.The UCSD CTEPH Quality Improvement Database requires significant retrospective review, and some factors that have been progressively found to be important in CTEPH, including echocardiogram measurements, metabolic markers, or type of anticoagulant, and adherence were not always recorded.Outcomes in the repeat PTE group may be confounded by selection bias, because the individuals who underwent repeat PTE were deemed by a multidisciplinary team to be optimal surgical candidates.Thus, their outcomes may not necessarily apply to all individuals with recurrent CTEPH.Because of the high degree of variance in the first PTE group, conclusions regarding postoperative complications cannot be directly due to repeat PTE status without controlling for multiple potential confounders.
In addition, because UCSD is a tertiary referral center, patients are often lost to follow-up after completing surgical The Journal of Thoracic and Cardiovascular Surgery c Volume 166, Number 6 intervention.Because of this, long-term outcomes, including follow-up hemodynamics, symptoms, use of pulmonary hypertension medications, health care use, and mortality, are not readily available.Moreover, surgical outcomes may not be generalizable to less-experienced centers.

CONCLUSIONS
In this large single-center series from an experienced CTEPH center, repeat PTE in select patients resulted in significant hemodynamic improvement with a low mortality, but increased postoperative complications, ventilator, ICU, and hospital days (Figure 4).This provides clarification regarding true surgical risk and supports referral for PTE, even if a patient has undergone prior PTE.In addition, this study illuminates potential causes of recurrent CTEPH including hypercoagulable state and anticoagulation failure, but human factors, including anticoagulation compliance, contribute significantly to recurrence.Finally, it is prudent to note that up to 25% of cases are due to inadequate resection during their first PTE, highlighting the importance of undergoing the evaluation and initial procedure at experienced, high-volume centers to avoid possible additional surgical morbidity and mortality.The Journal of Thoracic and Cardiovascular Surgery c December 2023 Thoracic: Lung Abbreviations and Acronyms BPA ¼ balloon pulmonary angioplasty CTEPH ¼ chronic thromboembolic pulmonary hypertension DOAC ¼ direct-acting oral anticoagulant ICU ¼ intensive care unit PTE ¼ pulmonary thromboendarterectomy PVR ¼ pulmonary vascular resistance UCSD ¼ University of California San Diego 46) The Journal of Thoracic and Cardiovascular Surgery c December 2023 Thoracic: Lung Astashchanka et al