A young adult patient who needed extracorporeal membrane oxygenation after pericardiocentesis due to post-pericardiotomy syndrome

Article information

Pediatr Emerg Med J. 2026;13(1):33-39

Publication date (electronic) : 2025 December 8

doi : https://doi.org/10.22470/pemj.2025.01431

Kyung Jin Ahn¹, Gi Beom Kim^,2

¹Department of Pediatric Cardiology, Gachon University Gil Medical Center, Incheon, Republic of Korea

²Department of Pediatrics, Seoul National University Children’s Hospital, Seoul National University College of Medicine, Seoul, Republic of Korea

Corresponding author: Gi Beom Kim Department of Pediatrics, Seoul National University Children’s Hospital, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 03080, Republic of Korea Tel: +82-2-2072-0266; E-mail: ped9526@snu.ac.kr

Received 2025 August 27; Revised 2025 October 2; Accepted 2025 November 6.

Abstract

Introduction

Pericardial decompression syndrome (PDS) is a rare but life-threatening complication of pericardiocentesis or surgical pericardiostomy, leading to paradoxical hemodynamic collapse and ventricular dysfunction. First described in 1983 (1), its exact cause remains unknown (2,3). Our case highlights acute ventricular collapse after pericardiocentesis, emphasizing the need for vigilant monitoring to detect and manage hemodynamic instability.

Case

A 27-year-old woman presented with a 2-week history of dyspnea and orthopnea to the emergency department (ED) at her 2 months postoperatively. Four months before the presentation, a routine checkup led to an echocardiographic diagnosis of complete atrioventricular septal defect with persistent primum atrial septal defect and spontaneously closed ventricular septal defect. Subsequently, she underwent atrial septal defect closure and mitral valve cleft repair without complications. At 1-month follow-up, she was asymptomatic, prompting diuretic discontinuation. Transthoracic echocardiography showed normal left ventricular (LV) systolic function, preserved right ventricular (RV) systolic function, only mild tricuspid regurgitation, and no evidence of pulmonary hypertension.

At the ED, her initial vital signs were as follows: blood pressure (BP), 97/69 mmHg; heart rate (HR), 120 beats/minute; respiratory rate, 24 breaths/minute; temperature, 36.7 °C; and oxygen saturation, 93% on room air. Heart sounds were distant, jugular venous pressure was elevated, with no definite pulsus paradoxus noted. Initial chest radiography showed cardiomegaly while electrocardiography showed a low QRS voltage (Figure 1). Echocardiography confirmed a large pericardial effusion with respiratory phase variation (Figure 2). The maximum depth of the effusion was approximately 45 mm.

Fig. 1.

Initial chest radiograph showing a cardiothoracic ratio of 0.73 (A) and electrocardiogram showing a low QRS voltage (B).

Fig. 2.

Initial transthoracic echocardiogram showing the mitral valve inflow Doppler pattern variation among respiration obtained at the emergency department.

In the ED, an urgent echocardiography-guided pericardiocentesis was performed within 2 hours of presentation, draining 200 mL of serosanguineous fluid, of which the analysis showed no signs of infection. Following the procedure, her vital signs remained as follows: BP, 90/66 mmHg; HR, 124 beats/minute; and respiratory rate, 31 breaths/minute. The rate of the pericardial fluid drainage was maintained at 100 mL/hour, with dobutamine infused at 3 μg/kg/minute.

Twelve hours later, after 1,200 mL had been drained, her systolic BP and urine output dropped to 70 mmHg and 0 mL for the 12 hours, respectively, while HR remained at 128 beats/minute. She also had worsening dyspnea, cool extremities, and signs of peripheral hypoperfusion, while her mental status remained alert with no cyanosis noted. At the time these symptoms manifested, an echocardiography was performed, which demonstrated severe biventricular dysfunction: LV ejection fraction < 10%, global hypokinesia, RV enlargement, and severe tricuspid regurgitation.

Cardiogenic shock persisted despite the volume loading, metabolic correction, and inotropic support. Given her worsening condition, venoarterial extracorporeal membrane oxygenation (ECMO) was initiated with LV venting (Figure 3A). On day 7, ECMO was successfully weaned, with cardiac function fully recovered. A follow-up radiograph showed a normalized cardiac silhouette (Figure 3B). She was discharged on day 23 and remained in good health at her 2-year follow-up.

Fig. 3.

Chest radiographs at the time of initiation of extracorporeal membrane oxygenation and at discharge. A chest radiograph on day 2, showing the cannulas in place following their insertion for the extracorporeal membrane oxygenation (A). A follow-up radiograph on day 23 at the time of discharge (B). RA: right atrium, LV: left ventricle.

Discussion

PDS is a rare but life-threatening complication of pericardiocentesis or surgical pericardiotomy, causing biventricular dysfunction and cardiogenic shock. This paradoxical hemodynamic instability occurs unexpectedly after treatment of cardiac tamponade. Despite being recognized for decades, the exact cause remains unknown with proposed mechanisms involving both catheter-based and surgical drainage (4).

Several hypotheses explain its pathophysiology (5,6). First, the hemodynamic hypothesis suggests that rapid pericardial drainage for treating cardiac tamponade increases venous return and overloads the LV. At the same time, the elevated systemic vascular resistance induced by the tamponade may persist transiently even after pericardiocentesis, creating a functional state similar to the ongoing tamponade. Second, the ischemic hypothesis states that compression of the coronary arteries by pericardial fluid results in myocardial ischemia, which may persist even after relief of compression due to a stunned myocardium. Third, the sympathetic withdrawal hypothesis proposes that compensatory sympathetic activation in the tamponade maintains cardiac output, but after rapid fluid removal, loss of this stimulus leads to cardiac dysfunction and collapse. A single unifying mechanism has yet to be identified, suggesting multiple contributing factors (7).

Incidence of PDS is reported to be 5%-11% following surgical pericardial drainage, while data on catheter-based pericardiocentesis remain limited (8). Mortality rates range from 30% to 80%, with surgical drainage carrying a higher risk (9). However, it is challenging to interpret these findings due to confounding factors.

One key risk factor is the rate of pericardial effusion withdrawal. Studies suggest that rapid removal of pericardial fluid increases the risk of cardiogenic shock (10-12). The underlying etiology of the pericardial effusion also affects prognosis. Clinicians should assess cytologic analysis for malignancy and evaluate myocardial function for underlying coronary artery disease or myocarditis. Following clinical stabilization, it may require coronary angiography, myocardial perfusion imaging, or cardiac magnetic resonance imaging.

Our patient was diagnosed with a complete atrioventricular septal defect at the age of 27 years and underwent the open-heart surgical repair. She later developed progressive pericardial effusion due to post-pericardiotomy syndrome. Given the risk of cardiac tamponade, pericardiocentesis was performed.

Differential diagnosis should include myopericarditis or infectious pericarditis (13). When pericardial effusion coexists with acute hemodynamic deterioration and ventricular dysfunction, it is essential to exclude acute inflammatory or infectious etiologies before attributing the collapse to PDS. In our patient, pericardial fluid analysis showed no evidence of infection, and peri-procedural cardiac enzymes did not exhibit a rise suggestive of acute myocarditis (Figure 4). Moreover, there were no clinical or laboratory features of systemic infection. Taken together with the temporal context after the recent cardiac surgery, these findings supported post-pericardiotomy syndrome complicated by PDS, rather than newly developed myopericarditis. While a subclinical inflammatory process can never be completely excluded, the preponderance of evidence favored PDS on the background of post-pericardiotomy effusion.

Fig. 4.

Temporal pattern of cardiac enzymes relative to key events (vertical bars). ECMO: extracorporeal membrane oxygenation, CKMB: creatine kinase-MB fraction.

To mitigate hemodynamic instability, the drainage rate was carefully controlled. At present, no standardized guidelines define the optimal rate or maximum safe volume for pericardial fluid removal during pericardiocentesis. However, despite cautious fluid removal, PDS still occurred. In our case, there was no prompt hemodynamic recovery immediately after the pericardiocentesis. Rather, her vital signs remained stable but not fully normalized while infusing low-dose dobutamine. Pericardial fluid was removed gradually at 100 mL/hour. Approximately 12 hours later, after a cumulative 1,200 mL had been drained, she developed cardiogenic shock with anuria. Transthoracic echocardiography at that time demonstrated severe biventricular systolic dysfunction. This clinical trajectory, lack of immediate improvement followed by delayed hemodynamic deterioration, aligns more closely with PDS than with classic tamponade or uncomplicated post-pericardiotomy syndrome. RV dysfunction was the primary manifestation, affecting LV systolic and diastolic function. Despite gradual drainage, she developed PDS, which can present with paradoxical hemodynamic collapse even after apparently successful pericardial fluid removal.

Venoarterial ECMO played a crucial role in stabilizing the patient by supporting end-organ perfusion while awaiting cardiac recovery. On day 7, the ECMO was successfully weaned off, and LV function fully recovered. For patients with low cardiac output syndrome, ECMO can serve as a lifesaving treatment (14).

PDS-related LV dysfunction must be differentiated from stress-induced cardiomyopathy (SIC), which presents as transient LV dysfunction with apical ballooning during systole. SIC mimics acute coronary syndrome, particularly anterior myocardial infarction. The classic “Takotsubo” shape of the LV is typically observed in elderly women following stress. Recent studies report cases involving mid-ventricular segments, excluding the apex. Unlike PDS, SIC generally resolves within 1 month (15).

Currently, no standardized preventive strategies exist for PDS. However, the rate and volume of pericardial fluid removal appear to be key modifiable factors. Gradual and controlled drainage allows adaptive hemodynamic compensation, reducing the risk of abrupt circulatory collapse and acute pulmonary edema following pericardial drainage (16,17) (Appendix, https://doi.org/10.22470/pemj.2025.01431).

Despite careful drainage control in our case, PDS still occurred. This underscores the importance of early recognition, continuous monitoring, and hemodynamic assessment. Given its potential severity, clinicians should be prepared to implement maximum pharmacologic support and, if necessary, ECMO as a rescue therapy for severe hemodynamic instability.

PDS is a rare yet life-threatening complication that may develop following pericardiocentesis, with its etiology still unknown. Since clinicians cannot completely prevent PDS due to multiple complex factors, increasing awareness of PDS to facilitate early treatment decisions and improve prognosis would be a more favorable strategy.

Article information Continued