Secondary Logo

Journal Logo

Anesthesia for Percutaneous Pulmonary Valve Implantation: A Case Series

Arboleda Salazar, Rafael MD*; Heggie, Jane MD*; Wolski, Piotr MD; Horlick, Eric MD; Osten, Mark MD; Meineri, Massimiliano MD*

doi: 10.1213/ANE.0000000000002904
Cardiovascular and Thoracic Anesthesiology: Original Clinical Research Report

BACKGROUND: Twenty percent of patients born with congenital heart disease present with right ventricular outflow tract abnormalities. These patients require multiple surgical procedures in their lifetime. Transcatheter pulmonary valve replacement (TPVR) has become a viable alternative to conventional pulmonary valve and right ventricular outflow tract surgery in pediatric and adult populations. In this retrospective review, we analyze the perioperative management of adult patients who underwent TPVR in our center.

METHODS: The study consisted of a chart review of patients who underwent TPVR at Toronto General Hospital between 2006 and 2015. Information about preoperative assessment, intraoperative anesthetic management, and intra- and postprocedural complications was collected. Two types of percutaneous valves have been used for a conduit or valve size between 16 and 28 mm. These procedures are done via the femoral, jugular, or subclavian vein under general anesthesia.

RESULTS: Seventy-nine adults (17–68 years of age) who underwent elective TPVR procedures were included. General anesthesia was used in all cases. Defibrillation was necessary in 1 case, and bradycardia was spontaneously resolved in another 1. Eighty-five percent were successfully extubated at the end of the procedure. Five patients required intraoperative inotropic support. Three patients presented self-resolved hemoptysis. Mechanical ventilation for >24 hours was necessary in 3 cases, 2 of which also required concomitant inotropic support. Four failed deployments and 1 case of persistent conduit stenosis were reported. Three patients required reintubation. All patients were discharged home.

CONCLUSIONS: Patients undergoing TPVR represent a complex and heterogeneous population. General anesthesia with endotracheal intubation is preferred. Setup for urgent lung isolation and cardiac defibrillation should be considered. Postoperative monitoring and intensive care setting are required. Anesthesiologists with cardiac anesthesia training are probably better suited to manage these patients.

From the *Department of Anesthesia, University Health Network, Toronto General Hospital, Toronto, Ontario, Canada

Department of Anesthesia, Heart and Chest Hospital, Liverpool, United Kingdom

Division of Cardiology, University Health Network, Toronto General Hospital, Toronto, Ontario, Canada.

Published ahead of print March 14, 2018.

Accepted for publication January 22, 2018.

Funding: None.

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Massimiliano Meineri, MD, Department of Anesthesia and Pain Management, Toronto General Hospital, 200 Elizabeth St, EN3-400, Toronto, ON M5G 2C4, Canada. Address e-mail to

Twenty percent of newborns with congenital heart disease present with right ventricular outflow tract (RVOT) defects.1 The most common defects associated with RVOT include abnormalities such as tetralogy of Fallot, pulmonary atresia, truncus arteriosus, transposition of the great arteries, and common arterial trunk.

These conditions require surgical corrections during childhood consisting of RVOT or pulmonary valve (PV) patch enlargements, PV replacement, or implantation of right ventricular (RV) to pulmonary artery (PA) conduit.2 RV–PA homografts are also used in the Ross procedure to correct aortic valve abnormalities. Prosthetic PV or conduit insufficiency and stenosis, patch dilation, or a combination of these are common midterm complications leading to RV dilation and dysfunction as well as atrial and ventricular arrhythmias, eventually resulting in heart failure with increased risk of sudden death. Optimal timing for intervention is debated.

Despite a relative low mortality, reoperative surgery is still associated with the increased morbidity of reoperative sternotomy and cardiopulmonary bypass, resulting in an increased transfusion requirement and need for postoperative intensive care unit (ICU) stay.3

Transcatheter PV replacement (TPVR)4 is a less-invasive alternative to open surgery for adult and teen patients with recurrent or native RVOT pathology. It was first described in the early 2000s,5 and its use has expanded worldwide as described by multiple trials and registries with a reported 90% technical success and procedural mortality between 0% and 1.6% (Table 1).6–14

Table 1.

Table 1.

TPVR has never been prospectively compared to open surgical PVR, but its short- and midterm results are comparable as well as its cost–benefit profile, despite a higher cost associated with the prosthetic valve and delivery system.15

The 2010 European guidelines support the use of TPVR with the same indications as surgical PVR16 and are summarized in Table 2. Nevertheless, the US guidelines still consider TPVR a class IIa level of evidence B (moderate-quality evidence) intervention.17

Table 2.

Table 2.

General anesthesia was used in most reported cases7,12,13; however, analysis of the specific anesthetic technique and intraoperative management for patients undergoing TPVR has not been described so far.

This report is needed because anesthesiologists should be aware of the potential problems when providing anesthesia for TPVR. Major periprocedural complications (Table 1) include the following: RVOT and PA tear and rupture, valvular embolization, and coronary artery compression.6–10,12,13,18,19

In this retrospective review, we analyze the perioperative management of adult patients who underwent TPVR in our center.

Back to Top | Article Outline


Our Institutional Research Ethics Board approved the study, and the requirement for written informed consent was waived. We retrospectively reviewed the electronic charts of a series of adult and teenage patients who underwent TPVR at Toronto General Hospital from July 2006 to December 2015 and were included in the institutional structural heart disease registry.

Patient preoperative characteristics were obtained from the cardiology clinical summary and the anesthetic preoperative assessment. All preoperative cardiac tests such as transthoracic echocardiography, cardiac magnetic resonance, and electrocardiogram were individually reviewed.

Scanned anesthetic charts were retrieved to define the anesthetic technique, type and dose of intravenous and inhalational anesthetics, intraoperative use of inotropes, airway management, duration of the procedure, and the time of tracheal extubation. Procedural reports were assessed to obtain information on the type of valve used, interventional approach, right heart catheterization data, arrhythmias, and procedural complications. Finally, ICU flow sheets and discharge letters were reviewed to define the need for postoperative inotropes and ventilatory support and ICU and hospital length of stay.

In our case series, the 2 currently available devices were used: The Melody (Medtronic Inc, Minneapolis, MN)5 and the Sapien transcatheter heart valves (Edwards Life Sciences, Irvine, CA).12 Both are balloon-expandable valves that are delivered via femoral, jugular, or subclavian venous approach through a 22F or 24F sheath. The Melody valve is a bovine internal jugular valved conduit sewn over a bare metal platinum–iridium stent with a height of 32 mm that can be expanded to a variable final diameter (from 16 to 22 mm). The Sapien valve is a bovine pericardial valve mounted on a cobalt chromium alloy frame with a height 14–16 mm available in 3 sizes (23, 26, and 29 mm) to achieve a final diameter (from 22 to 28 mm).

Device choice is based on PV diameter measured intraoperatively and center-specific experience.

The TPVR procedure consists of 3 steps: RVOT balloon dilation, stenting of the landing zone, and valve deployment. Balloon dilation is performed with a large balloon inflated with radioopaque contrast under fluoroscopy. It serves the purposes of measuring the size of the conduit or annulus for the new valve, identifying an increased risk of rupture (in case of a pronounced waist indentation), and dilating the conduit. Given the increased incidence of abnormal coronary artery anatomy in this patent population, coronary injection is performed at the time of balloon inflation to rule out coronary compression. Balloon dilation temporarily causes complete RVOT occlusion. After dilation, a guidewire is advanced into a distal PA branch and used for all subsequent procedure steps. RVOT stenting provides a homogenous landing zone for the new valve prosthesis and stiffens the RVOT, avoiding possible recoil during deployment, preventing rupture, and ensuring adequate seating. Stenting is not necessary for the Melody valve given its height, but it is always necessary for the low-profile Sapien valve. Stenting causes acute severe pulmonary insufficiency (PI).

Complications were defined as any threatening event requiring interventional treatment or a life-threatening event requiring life support.

Back to Top | Article Outline

Statistical Analysis

Median and range and mean ± standard deviation were calculated for continuous variables. Numbers and percentages were calculated for categorical variables.

Back to Top | Article Outline




We identified 79 adults (17–68 years of age) who underwent TPVR at our institution between July 2006 and December 2015. There was an upward trend for number of procedures per year with a simultaneous decrease of the mean procedure duration (Figure).

Back to Top | Article Outline

Preoperative Characteristics

The patients’ demographics, characteristics, comorbidities, procedures, indications, ventricular functions, and RVOT gradients are shown in Table 3.

Table 3.

Table 3.

The indication for TPVR varied significantly among our group and included PV insufficiency, stenosis, or RVOT stenosis. It was related to degeneration and malfunction of previously implanted bioprosthesis, RV to PA conduit, RVOT homograft, or primary disease of the native valve or the RVOT. In 2 cases, stenotic RVOT or PV lesion was stented ahead of time to allow stent reepithelialization leaving the patient with severe PI. None of the patients presented for emergent TPVR or with refractory heart failure.

Back to Top | Article Outline

Anesthesia Management

All the procedures were performed under general anesthesia (Table 4). A combination of midazolam (0.02 mg/kg), propofol (2 mg/kg), fentanyl (2 µg/kg), and rocuronium (5 mg/kg) was used for induction in all cases. Addition of ketamine (0.5 mg/kg) was reported in 7 cases (8%). Sevoflurane (minimal alveolar concentration between 0.5 and 1.2) was the anesthetic vapor of choice for inhalational anesthesia (69), while remifentanil infusion (0.05–0.25 µg/kg/min) was used in 10 cases. Total intravenous anesthesia with a continuous infusion of propofol (75–150 µg/kg/min) and remifentanil (0.05–0.25 µg/kg/min) was used in 3 cases.

Table 4.

Table 4.

One patient had a supraglottic airway device; the remaining 78 underwent endotracheal intubation with a single-lumen tube. One patient required complex manipulation of the airway and needed elective tracheal dilation before intubation for known tracheal stenosis. In all patients, anesthetic was managed toward tracheal extubation at the end of the procedure.

Standard anesthesia monitoring was used, consisting of invasive and noninvasive arterial pressure monitoring, electrocardiography, peripheral capillary oxygen saturation, end-tidal carbon dioxide, and respiratory gas analysis. A radial arterial line was inserted under local anesthesia before induction.

A large-gauge peripheral intravenous line (14 or 16 gauge) was placed either before or immediately after induction of anesthesia on either forearm. An upper body forced-air warming blanket was used in all cases. Central venous access was secured connecting to a sterile tube extension at the side port of the femoral or jugular venous valve delivery sheath. None of the patients had a central venous catheter inserted by the anesthesiologist.

All patients were positioned supine with the arms above the head and the elbows secured to allow laterolateral fluoroscopy. Eyes were padded, and 2 pillows were used to protect the face and avoid excessive arm extension.

Precautions as for redo sternotomy were taken in all cases. They included external defibrillator pads, 2 units of crossed-matched blood available, and full-dose heparin drawn for possible urgent initiation of cardiopulmonary bypass and ready to be administered.20,21 Radioopaque defibrillator pads were positioned on the right scapula and below the right nipple so as not to interfere with fluoroscopic imaging and connected to the defibrillator.

Systemic heparinization was achieved in all cases to maintain an activated clotting time of >300 seconds and prevent clot formation on catheters and wires. The initial heparin dose was 100 IU/kg.

Rapid ventricular pacing was performed in 3 cases to aid valve deployment. It was achieved using a temporary pacemaker wire inserted through the femoral vein using a separate 6F introducer.

None of the patients required blood transfusions. All patients only received crystalloids as intravenous fluids maintenance, and the average fluid balance was 832 ± 591 mL.

Intraoperative infusion of milrinone (0.25–0.5 µg/kg/min) was used in 3 cases, in 1 case in combination with epinephrine (0.05 µg/kg/min) and norepinephrine (0.05 µg/kg/min). Dobutamine (3 µg/kg/min) was used in 1 case, and norepinephrine (0.05–0.1 µg/kg/min) was used in 7 cases (Table 4).

Sixty-seven patients (85%) were successfully extubated in the operating room at the end of the procedure; the remaining 12 were transferred intubated to the ICU and 9 of them extubated within 24 hours.

Back to Top | Article Outline


The Melody valve was used in 50 cases and the Sapien valve in 29 cases. The femoral venous approach was used in 76 cases and the right internal jugular in 3 cases. In 1 of the latter, the PV was accessed via an in situ Glenn shunt. Fluoroscopy was the only imaging modality used to guide the procedure.

The TPVR prosthesis was deployed in a valved conduit in 22 cases, in a prosthetic valve in 33 cases, in a homograft in 19 cases, in a previously delivered stent in 2 cases, and in the native PV in 3 cases.

Deployment of a second valve was necessary in 2 cases after Sapien valve TPVR: in 1 case after prosthesis embolization and subsequent retrieval, and in the second case for severe residual paravalvular leak. Percutaneous venous access closure devices were used in all cases.

Back to Top | Article Outline


No perioperative deaths were reported; however, several complications were identified in the perioperative period (Table 5).

Table 5.

Table 5.

In 4 cases, the procedure was aborted: in 3 cases, the PV annulus was too large for the Melody valve, and in the other case, the delivery catheter was not long enough to reach the deployment site.

Valve embolization was reported in 1 case and required surgical femoral cut down to retrieve it. In 1 case, significant residual gradient across the RVOT required subsequent elective open PV replacement 2 days later.

Puncture site hematomas developed in 2 patients and were complicated by infection or pseudoaneurysm.

Ventricular tachycardia requiring cardioversion occurred in 3 cases before valve deployment and cardioversion was required.

Right bundle branch block was prevalent in our patient population (69 patients) and was associated with left fascicular block in 3 cases. Nevertheless, the incidence of intraoperative bradycardia was extremely low. Transient third-degree atrioventricular block was observed in 1 single case after balloon dilation.

Three patients required tracheal reintubation in the first 24 hours for respiratory failure.

Hemoptysis was reported in 4 cases. In 3 cases, it occurred intraoperatively after valve deployment and in 1 case in the ICU at the time of extubation. Spontaneous resolution after tracheal reintubation with a double-lumen tube was observed in all cases. Intraoperative hemoptysis was associated with a moderate-size left hemothorax (400 mL) in 1 case, drained successfully with a left-sided chest tube without need for any further intervention.

The 3 cases of tracheal reintubation within 24 hours from the procedure were related to aspiration pneumonia, hemoptysis, and pulmonary edema, respectively.

In the immediate postoperative period, all patients were monitored in cardiac ICU overnight and discharged to a regular floor bed when stable.

Three patients required mechanical ventilation for >24 hours. Signs of pulmonary congestion on chest x-ray were present in all cases and improved with diuretics. In 1 case it was associated with ventricular failure and in another with aspiration pneumonia.

Two patients were admitted to the ICU on inotropes: milrinone for left ventricular failure and milrinone and epinephrine for biventricular failure. The patient with biventricular failure required reintubation in the first 24 hours and developed acute renal failure requiring dialysis.

Hospital and ICU length of stay are summarized in Table 5. Patients with perioperative complications had prolonged length of stay, with an average of 4.3 ± 3.5 days in the ICU and 8.9 ± 10.1 days in the hospital.

Back to Top | Article Outline


Our case series highlights the complexity and heterogeneity of patients undergoing TPVR.

Similarly to previously published series,6–10,12,13 ours included mostly young patients with congenital heart disease and multiple previous procedures as well as a group of patients with failed RVOT homograft conduit after Ross procedure (Table 3). In 36 of our cases, TPVR was deployed off-label in native and prosthetic PV as TPVR is currently only approved for RVOT or conduit defects.22

RVOT tear6–8,10,12,18 was not uncommon in previously published reports and was successfully treated with a covered stent without resulting into major morbidity in most cases.13 Scarring from previous surgeries and the presence of an RVOT stent or bioprosthetic valve may have played a protective role in preventing RVOT rupture during balloon dilation and valve deployment in our cohort.

Our case series did not include any emergent deployment in patients with preoperative decompensated heart failure that resulted in poor short-term outcomes in other reports.8

All our patients underwent general anesthesia. The choice of anesthetic drugs was left to the attending anesthesiologist, and it was varied and could not be correlated to specific patient characteristics. In previous case series, general anesthesia was more commonly used7,9,11–13 than sedation.9,11 However, the type of anesthesia (inhalational versus total intravenous anesthesia) and specific medication used were not reported before (Table 4). Some authors referred to the use of “heavy sedation”9,11 without defining it.

The choice of general anesthesia is mostly driven by 3 factors: duration of the procedure (average of 204 ± 99.2 minutes of total room time), patient’s uncomfortable position, need to avoid coughing and movements during deployment.

Endotracheal intubation with a standard tube was used in all cases except 1. Given the patient’s position and the need for access to the airway, we believe that endotracheal intubation with careful fixation of the endotracheal tube is a critical safety concern. Extension tubing is also necessary in all cases to avoid accidental extubation during the movement of the fluoroscopy equipment.

Baseline and postdeployment right heart catheterization including chambers and PA pressures and cardiac output measurement were the only objective indices of RV function intraoperatively. Use of transesophageal echocardiography has been reported by other groups6 and would have allowed ideal continuous hemodynamic assessment. RV support with milrinone and/or epinephrine was started empirically in some patients with low baseline cardiac output or to treat postdeployment hypotension in patients with known RV dysfunction. The use of vasopressors (Table 4) may be in part justified by the negligible surgical stimulation during general anesthesia. Inotropic support was discontinued in all cases based on improved hemodynamics after valve deployment except in 2. In 1 case of severely reduced biventricular function at baseline with severe PI, the patient did not experience any hemodynamic improvement after TPVR, and in the second case, the patient developed acute LV dysfunction after TPVR despite normal biventricular function at baseline, likely as a result of increased LV preload after relief of severe pulmonary stenosis.

Complete atrioventricular block was observed only once and self-resolved, and ventricular tachycardia requiring electrical cardioversion was observed in 2 patients.

Hemoptysis is the result of PA rupture or guidewire injury. Intraprocedural and delayed hemoptysis were reported by others but never required specific treatment. We suggest that a double-lumen tube and a tube exchange catheter should be available in the procedure room in case of hemoptysis.23

This study has a number of limitations. It is a small single-center retrospective case series including young patients only. Comparison with previously reported case series is limited by the fact that pediatric patients were often included in other reports. Second, 2 different types of valve prostheses were used when a single type was included in previous reports. Finally, it is the only one to our knowledge in which the 2 available devices were used.

Our review highlights some of the intraoperative anesthetic challenges for TPVR. Although the ideal anesthetic regimen cannot be defined, we believe that general anesthesia with endotracheal intubation provides a controlled environment to manage the frequent possibly life-threatening complications such as hemoptysis and arrhythmias. We identified a significant heterogeneity within patients who developed complications (Table 5), making their prediction very difficult. The potential need for prolonged ventilation and reintubation mandates postoperative monitoring and ICU backup.

Back to Top | Article Outline


Name: Rafael Arboleda Salazar, MD.

Contribution: This author helped perform the chart review, do the literature search, create the outline, and write the first manuscript draft.

Name: Jane Heggie, MD.

Contribution: This author helped review the first draft and subsequent revisions.

Name: Piotr Wolski, MD.

Contribution: This author helped perform the chart review, participate in creating the outline, and review the first draft and subsequent revisions.

Name: Eric Horlick, MD.

Contribution: This author helped review the first draft and subsequent revisions.

Name: Mark Osten, MD.

Contribution: This author helped review the first draft and subsequent revisions.

Name: Massimiliano Meineri, MD.

Contribution: This author helped conceive this study, create the outline, and edit the original manuscript and subsequent revisions.

This manuscript was handled by: Nikoloas J. Skubas, MD, DSc, FACC, FASE.

Back to Top | Article Outline


1. Warnes CA, Williams RG, Bashore TM, et al.ACC/AHA 2008 guidelines for the management of adults with congenital heart disease: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to develop guidelines for the management of adults with congenital heart disease): developed in collaboration with the American Society of Echocardiography, Heart Rhythm Society, International Society for Adult Congenital Heart Disease, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. Circulation. 2008;118:2395–2451.
2. Murphy JG, Gersh BJ, Mair DDLong-term outcome in patients undergoing surgical repair of tetralogy of Fallot. N Engl J Med. 1993;329:593–599.
3. Ong K, Boone R, Gao M, Kiess M, Grewal JRight ventricle to pulmonary artery conduit reoperations in complex congenital heart disease: what is the optimal timing of transcatheter pulmonary valve implantation? Can J Cardiol. 2012;28:S406–S407.
4. Ansari MM, Cardoso R, Garcia DPercutaneous pulmonary valve implantation: present status and evolving future. J Am Coll Cardiol. 2015;66:2246–2255.
5. Bonhoeffer P, Boudjemline Y, Saliba ZPercutaneous replacement of pulmonary valve in a right-ventricle to pulmonary-artery prosthetic conduit with valve dysfunction. Lancet. 2000;356:1403–1405.
6. Armstrong AK, Balzer DT, Cabalka AKOne-year follow-up of the Melody transcatheter pulmonary valve multicenter post-approval study. JACC Cardiovasc Interv. 2014;7:1254–1262.
7. Butera G, Milanesi O, Spadoni IMelody transcatheter pulmonary valve implantation: results from the registry of the Italian Society of Pediatric Cardiology. Catheter Cardiovasc Interv. 2013;81:310–316.
8. Cheatham JP, Hellenbrand WE, Zahn EMClinical and hemodynamic outcomes up to 7 years after transcatheter pulmonary valve replacement in the US melody valve investigational device exemption trial. Circulation. 2015;131:1960–1970.
9. Eicken A, Ewert P, Hager APercutaneous pulmonary valve implantation: two-centre experience with more than 100 patients. Eur Heart J. 2011;32:1260–1265.
10. Fraisse A, Aldebert P, Malekzadeh-Milani SMelody® transcatheter pulmonary valve implantation: results from a French registry. Arch Cardiovasc Dis. 2014;107:607–614.
11. Haas NA, Moysich A, Neudorf UPercutaneous implantation of the Edwards SAPIEN(™) pulmonic valve: initial results in the first 22 patients. Clin Res Cardiol. 2013;102:119–128.
12. Kenny D, Hijazi ZMTranscatheter approaches to non-valvar structural heart disease. Int J Cardiovasc Imaging. 2011;27:1133–1141.
13. Lurz P, Coats L, Khambadkone S, et al.Percutaneous pulmonary valve implantation: impact of evolving technology and learning curve on clinical outcome. Circulation. 2008;117:1964–1972.
14. McElhinney DB, Hennesen JTNyas. The Melody(R) valve and Ensemble(R) delivery system for transcatheter pulmonary valve replacement. Pharmaceutical Science to Improve the Human Condition: Prix Galien USA 2012. 2013:Oxford: Blackwell Science Publishers, 77–85.
15. Borik S, Crean A, Horlick E, et al.Percutaneous pulmonary valve implantation: 5 years of follow-up: does age influence outcomes? Circ Cardiovasc Interv. 2015;8:10.
16. Baumgartner H, Bonhoeffer P, De Groot NMTask Force on the Management of Grown-up Congenital Heart Disease of the European Society of Cardiology (ESC); Association for European Paediatric Cardiology (AEPC); ESC Committee for Practice Guidelines (CPG). ESC guidelines for the management of grown-up congenital heart disease (new version 2010). Eur Heart J. 2010;31:2915–2957.
17. Feltes TF, Bacha E, Beekman RH 3rdAmerican Heart Association Congenital Cardiac Defects Committee of the Council on Cardiovascular Disease in the Young; Council on Clinical Cardiology; Council on Cardiovascular Radiology and Intervention; American Heart Association. Indications for cardiac catheterization and intervention in pediatric cardiac disease: a scientific statement from the American Heart Association. Circulation. 2011;123:2607–2652.
18. McElhinney DB, Hellenbrand WE, Zahn EMShort- and medium-term outcomes after transcatheter pulmonary valve placement in the expanded multicenter US melody valve trial. Circulation. 2010;122:507–516.
19. Mihăilă S, Muraru D, Piasentini EQuantitative analysis of mitral annular geometry and function in healthy volunteers using transthoracic three-dimensional echocardiography. J Am Soc Echocardiogr. 2014;27:846–857.
20. Maciejewski M, Piestrzeniewicz K, Bielecka-Dąbrowa A, Piechowiak M, Jaszewski RRedo surgery risk in patients with cardiac prosthetic valve dysfunction. Arch Med Sci. 2011;7:271–277.
21. Machiraju VRRedo Cardiac Surgery in Adults. 2012.2nd ed. New York, NY: Springer;
22. Hascoët S, Acar P, Boudjemline YTranscatheter pulmonary valvulation: current indications and available devices. Arch Cardiovasc Dis. 2014;107:625–634.
23. Damm C, Degen H, Stoepel C, Haude M[Management of a catheter-induced rupture of a pulmonary artery]. Dtsch Med Wochenschr. 2010;135:1914–1917.
Copyright © 2018 International Anesthesia Research Society