Emergency Veno-Arterial Extracorporeal Membrane Oxygenation for Pericardial Decompression Syndrome

Background Pericardiocentesis is a therapeutic lifesaving intervention for patients presenting with cardiogenic shock due to pericardial effusion with signs of tamponade. Pericardial decompression syndrome (PDS) is a rare fatal complication that may occur after pericardiocentesis. Case Presentation. We report a case of a patient with idiopathic primary pulmonary hypertension who presented with massive pericardial effusion complicated with rapid hemodynamic and respiratory deterioration. Gradual therapeutic pericardiocentesis was done but progressive circulatory collapse occurred. Emergent veno-arterial extracorporeal membrane oxygenation (VA-ECMO) was applied. Echocardiography revealed severe right ventricle failure. Unfortunately, the patient developed acute progressive thrombocytopenia and bilaterally diffuse subarachnoid hemorrhage after 4 days of ECMO support. Conclusions Therapeutic pericardiocentesis can be occasionally fatal in cases of significant pulmonary hypertension with massive pericardial effusion when complicated by pericardial decompression syndrome. Acute significant thrombocytopenia may occur with VA-ECMO support resulting in fatal bleeding.


Background
Idiopathic pulmonary artery hypertension (IPAH) is a rare rapidly progressive disease affecting the pulmonary precapillary vasculature and results in right-sided heart failure and death. The World Health Organization (WHO) classified IPAH as a part of group I pulmonary hypertension [1,2]. The increased pulmonary artery resistance is related to endothelial dysfunction with vasoconstriction, remodeling, and thrombosis. The endothelial dysfunction results from imbalance between different vasoactive substances that affect intracellular nitric oxide, endothelin, and prostacyclin pathways [3,4].
Pericardiocentesis is a therapeutic lifesaving intervention for patients presenting with cardiogenic shock due to pericardial effusion with signs of tamponade [5,6]. Pericardial decompression syndrome (PDS) is a rare fatal complication that may occur after pericardiocentesis. It is defined as paradoxical hemodynamic deterioration after successful pericardiocentesis. It was referred to that condition as paradoxical hemodynamic instability or low cardiac output syndrome [7,8].

Case Presentation
The patient was a 28-year-old female patient with a body mass index of 21.3 (kg/m 2 ). She was diagnosed to have idiopathic pulmonary artery hypertension (IPAH) for 3 years and was maintained on maximized pulmonary vasodilators including oral riociguat and macitentan and intravenous continuous infusion of treprostinil. She started to develop progressive right-sided heart failure including bilateral lower limb edema, dyspnea on minimal effort, and pleural and pericardial effusions. She was listed for lung transplantation. She was admitted to our institution with 1 week of exaggerated dyspnea and palpitations. On admission, she presented with sinus tachycardia of 130 beats/min, blood pressure of 90/45 mmHg with presence of jugular venous distention, and pulse oximetry saturation (SPO2%) of 80% on room air. After central venous catheterization, CVP range was 25-30 cm H 2 O (Figure 1).
Chest X-ray revealed increased cardiothoracic ratio with increased interstitial thickenings without lung collapse nor consolidation. Laboratory work-up revealed white blood cell count of 6.36 (10 9 /L), platelet count of 189 (10 9 /L), hemoglobin 114 (g/L), serum Na 132 mmol/L, NT-proBNP 3693 pg/mL, normal kidney, and liver chemistries. Septic and virology screening was done including COVID-19 PCR. Echocardiography was done and revealed a large circumferential pericardial effusion without signs of tamponade. Also it showed a severe dilatation of right ventricle with severe systolic dysfunction and flattened interventricular septum. There was a severe right atrium dilatation with left sided shift of interatrial septum. The estimated pulmonary artery systolic pressure (PASP) was more than 115 mmHg while the left ventricle was underfilled but with a good systolic function ( Figure 2).
After hemodynamic stabilization with inotropic support and use of high flow nasal oxygen, intravenous frusemide was used to decrease the volume overload as indicated with the generalized anasarca. After debate about the possible benefit and risks of pericardiocentesis, the patient was taken to the catheterization laboratory where pericardiocentesis was done under fluoroscopic and echocardiographic guidance with immediate aspiration of 250 mL pericardial serous fluid. The pericardial effusion was transudate without any pathogen. The analysis revealed the following: fluid protein 38 g/L, albumin 26 g/L, LDH 177 units/L, triglycerides 0.3 mmol/L, no bacteria detected with Gram stain and culture, and no acid fast bacilli detected.
Despite gradual withdrawal of 1.3-liter pericardial fluid over the next 24 hours, the patient developed hemodynamic deterioration with progressive lactic acidosis and rising NT-proBNP to 9299 (pg/mL). Emergency echocardiography revealed normal left ventricle while severely dilated right ventricle with systolic dysfunction and increased PASP more than 150 mmHg and minimal pericardial effusion ( Figure 3).
Emergent peripheral veno-arterial extracorporeal membrane oxygenation (VA-ECMO) via femoral approach with reperfusion cannula was applied and cardiopulmonary resuscitation was done. Anticoagulation started with unfractionated heparin infusion guided by activated clotting time (ACT) target of 180-220 seconds, heparin level (target 0.3-0.7 units/mL), and antithrombin III activity (target > 50 %).The patient had acute kidney injury with decrease of glomerular filtration rate (GFR) from more than 60 to 40 mL/ min/1.73 m 2 , but gradual recovery happened without renal replacement therapy. Neurological assessment was frequently done during sedation withdrawal and pupils' size and reactivity to light every 2 hours according to our hospital protocol. Near-infrared spectroscopy (NIRS) was used continuously for monitoring of cerebral and lower limb oxygen saturations. The patient developed a rapid decline of platelet count, and heparin-induced thrombocytopenia (HIT) was unlikely. Despite the HIT 4T score was 6, the heparin-PF4 antibody test was 0.11 (reference ≤ 0:39). Exclusion of significant hemolysis was done by laboratory workup: blood schistocyte was <0.5%, Coombs test was negative, haptoglobin level was 0.7 (reference: 0.3-2 g/L), peak bilirubin was 42.4 (reference: 0-21 μmol/L), and LDH 585 (reference: 135-214 units/L). The patient developed a thrombus at the tip of ECMO drainage cannula which made the decision to stop anticoagulation exceedingly difficult. So frequent platelet transfusions were done to keep platelet count more than 50 (10 9 /L). Daily sedation withdrawal and neurological assessment was done, and continuous monitoring of brain oxygen saturation by near-infrared spectroscopy (NIRS) was maintained.
After 4 days of ECMO support, the patient developed accelerated systemic hypertension after sedation withdrawal and deterioration of consciousness; urgent brain computed tomography (CT) imaging revealed left occipital hemorrhagic stroke and bilateral diffuse subarachnoid hemorrhage with intraventricular extension complicated with hydrocephalus. Cerebral angiography excluded aneurysmal dilatation or arteriovenous cerebral malformation. Relief of intracranial hypertension was done using external ventricular draining, head positioning, keeping normocapnia, and osmotic diuresis. Anticoagulation was discontinued immediately when the consciousness was impaired, and platelet count was kept more than 100 (10 9 /L). After 4 days of the cerebral bleeding, the patient developed brain death and brain CT imaging revealed diffuse brain ischemia. Withdrawal of support was done, and the patient was declared dead (Figure 4).

Discussion
Pericardiocentesis is a therapeutic lifesaving intervention for patients presenting with cardiogenic shock due to pericardial tamponade. It is safe, and the risk of complications ranges from 4% to 10% including coronary artery or cardiac chamber puncture, arrhythmias, hemothorax, and pneumothorax [5,6]. Pericardial decompression syndrome (PDS) is a rare fatal complication that may occur after pericardiocentesis. It is defined as paradoxical hemodynamic deterioration after successful pericardiocentesis. It was referred to that condition as paradoxical hemodynamic instability or low cardiac output syndrome [7,8]. Few cases were reported with PDS presented with uni-or biventricular dysfunction [9][10][11][12][13][14]. Also, few cases were reported with acute pulmonary edema even with normal ventricular function [15][16][17]. Our patient had already a right ventricular (RV) dysfunction that was aggravated after pericardiocentesis even with slow cautious drainage. After drainage, she had severe right ventricle failure with normal left ventricular function resulting in cardiogenic shock and respiratory failure. Emergent VA-ECMO was applied for rapid resuscitation. Emergency pericardiocentesis is a class IA recommendation for unstable patients and preferably with fluoroscopic and echocardiographic guidance which was done in our patient [5].
Pradhan et al. [18] did analysis of the 35 reported cases of PDS and found the volume of drained fluid ranged from 450 to 2,100 mL, and the onset of hemodynamic deterioration varied from immediate to 48 hours after pericardiocentesis. Most of the presentations were left ventricular failure and pulmonary edema, while RV failure was less frequent 2 Case Reports in Cardiology presentation. Most of the reported cases of PDS died between 6 hours and 14 days after pericardiocentesis. There are many proposed mechanisms of PDS including rapid drainage of pericardial fluid with rapid increase of venous return and ventricular overloading [9,10,17,19]. In our case, the drainage was slowly done and achieved over 24   3 Case Reports in Cardiology hours, but there are no definite guidelines for the proper amount to be drained, especially with preexisting pulmonary hypertension and RV dysfunction. Another proposed mechanism was myocardial stunning and systolic dysfunction due to reduced coronary perfusion pressure [10,12,18,20,21]. Our echocardiography after deterioration showed normal LV function without regional wall motion abnormalities.
After failure of inotropic support and high flow nasal oxygen to achieve stabilization, rescue VA-ECMO was applied via femoral approach. The use of VA-ECMO is still associated with high mortality and many morbidities including neurological and vascular injuries, bleeding, and thrombocytopenia which were frequently studied [22][23][24][25][26][27][28][29]. The patient developed progressive thrombocytopenia without bleeding, and frequent platelet transfusions were given. Echocardiography was repeated and showed a thrombus at the tip of drainage cannula but did not affect the ECMO flow. Heparin-induced thrombocytopenia (HIT) is a serious problem that can happen after few days of exposure to heparin and associated with arterial and venous thrombotic complications and rarely bleeding [30,31]. We calculated the 4T score and it showed high probability, but the serology rejected the diagnosis of HIT.
The 4Ts is a scoring for probability of HIT and incorporates 4 items including thrombocytopenia magnitude, timing after exposure to heparin, thrombotic events, and possible other causes of thrombocytopenia. The scores of 0-3, 4-5, and 6-8 were considered as low, intermediate, and high probability for HIT, respectively [32,33]. Cuker et al. [34] conducted a meta-analysis and concluded that a low probability 4Ts score excludes HIT without need for laboratory testing, while with intermediate or high scores, heparin should be discontinued and laboratory testing should be requested. Moreover, only 7-12% of patients with suspected HIT referred for laboratory testing had positive results [35,36].
A recent meta-analysis of 12 studies reported the prevalence of thrombocytopenia in 23.2% (95% CI 11.8-34.5; 6 studies) and occurrence of platelet dysfunction during VA-ECMO support without association with ECMO duration [37]. Jiritano et al. [37] reported the decline of platelet count during first 7 days of ECMO initiation and proposed a multifactorial theory including contact with the extracorporeal circuit, inflammatory  The occurrence of intracerebral bleeding was sudden and significant that necessitated neurosurgical intervention to insert external ventricular drain and decrease the intracranial pressure. Female gender, thrombocytopenia, and low body mass index were linked to early intracerebral bleeding in ECMO patients [26,27,40].
Finally, our case highlights the importance of gradual judicious decompression during therapeutic pericardiocentesis especially in presence of significant pulmonary hypertension to avoid acute right ventricle failure or pericardial 5 Case Reports in Cardiology decompression syndrome. Also, careful hemodynamic monitoring after pericardiocentesis should be done for early detection of impaired tissue perfusion and need for cardiopulmonary support. Moreover, acute significant thrombocytopenia may happen shortly after ECMO support and may be complicated with fatal bleeding.

Conclusions
Therapeutic pericardiocentesis can be occasionally fatal in cases of significant pulmonary hypertension with massive pericardial effusion when complicated by pericardial decompression syndrome. Acute significant thrombocytopenia may occur with VA-ECMO support resulting in fatal bleeding.

Data Availability
The data of the case study is available from the corresponding author.

Ethical Approval
The case study was approved by the ethical committee of King Faisal Specialist Hospital and Research Center (KFSHRC).

Consent
The authors got approval to publish this case study by the ethical committee of King Faisal Specialist Hospital and Research Center without a consent from the patient's family as there is no any personal information that could lead to the identification of the patient, and deidentification was done in accordance with the Health Insurance Portability and Accountability Act of 1996 (HIPAA) Privacy Rule (https:// www.hhs.gov/hipaa/for-professionals/privacy/specialtopics/de-identification/index.html).

Conflicts of Interest
The authors declare that no competing interest.

Authors' Contributions
ML participated in data collection, interpretation, and drafting of the manuscript. RZ participated in data interpretation and writing of the manuscript. MG and PM participated in data collection. AA participated in data collection and interpretation of echocardiography images. All authors read and approved the final manuscript for publication.