Case Report: Pericardial effusion in late onset neonatal Escherichia coli sepsis

Introduction

Pericardial effusion (PE) is a rare condition in neonates. Symptoms include difficulty breathing and fever. Without appropriate treatment, it can quickly progress to hemodynamic collapse, tamponade, and ultimately, death. The most common cause of pericardial effusion in neonates are central venous catheters. It is due to the mechanical and osmotic injury of the thin and immature myocardium in neonates and especially preterm.¹

Case report

A preterm female infant and her twin sister were born at 34 weeks of gestation via an emergency caesarean section due to covid 19 infection in the mother. Maternal serologies (toxoplasmosis, rubella, hepatitis B and syphilis) and prenatal ultrasound findings were unremarkable.

The first twin was small for gestational age and had severe respiratory distress. She was referred to the PICU. She was discharged on day 10 of hospitalization. No further symptoms were noted.

Our patient weighted 1600 g (small for gestational age), and Apgar score was 8 at 1 min and 9 at 10 min. She was referred to out neonatal unit in the pediatric ward for respiratory distress. At presentation, respiratory rate was 62 per minute, heart rate was 138 per minute and oxygen saturation was 97%. Chest radiography was normal. Covid 19 polymerase chain reaction test was negative. She was placed on oxygen therapy via high-flow nasal cannula for three days. The patient received fluids through a peripheral venous catheter. The diagnosis of early onset neonatal infection was discussed but excluded based on: rapid improvement of respiratory status, normal CRP levels and negative blood culture. The patient did not receive any antibiotics. Final diagnosis was transient tachypnea of the newborn. As her respiration improved, she was placed on room air on day 3 of life, fluid infusion was discontinued on day 4 of life, and she was gradually fed with milk formula (breast milk was not available as her family lived from our facility). She was fed initially via gastric tube and gradually via bottle. The patient was discharged on day 14 of life.

On day 18 of life, she presented to the emergency room with fever and dyspnea. Respiratory rate was 65/minute with intercostal recession. Oxygen saturation was 97% in room air. The patient was placed on oxygen via a nasal cannula. Chest radiography was normal. Laboratory tests showed a leukocyte count of 26000/mm³ and an elevated reactive protein C level of 220 mg/l. Initially, a nosocomial infection was suspected, and the patient was received imipenem and amikacin. There was no improvement in her respiratory status.

On day 21 of life, two blood cultures obtained on the day of readmission were positive for multidrug-sensitive Escherichia coli (E. coli). The patient was then switched to cefotaxime. On day 23 of life, our patient developed signs of heart failure with tachycardia, hepatomegaly, tachypnea, and a capillary refill of 4 seconds. Chest radiography revealed cardiomegaly (cardiothoracic ratio 0.67) with a globular heart shape (Figure 1). Electrocardiogram was normal. Cardiac ultrasound was not available in our hospital so it was performed by cardiologists in another facility. It revealed echogenic, circumferential pericardial effusion (11 mm) and pre tamponade (protodiastolic right ventricle collapse). It also showed a small patent ductus arteriosus. There were no signs of endocarditis.

Figure 1. Pre-pericardiocentesis chest radiography demonstrating marked cardiomegaly with globular cardiac silhouette.

A pericardial drain was inserted surgically, and 8 ml of purulent fluid mixed with blood was aspirated. Cytological analysis showed 3 leukocytes per ml. Biochemical analysis was not performed. No bacteria were isolated in pericardial fluid. After pericardial drainage, the respiratory status improved, and the shape of the heart was normal on chest radiography (Figure 2). The pericardial catheter was removed 10 days later. Cefotaxime was continued for a total duration of 14 days. A second cardiac ultrasound was performed at the age of 32 days. It showed no pericardial effusion. The patient was discharged on day 35 of hospitalization. At discharge, her weight was 2550 g and her physical examination showed normal body temperature, respiratory rate 34 per minute and heart rate 124 per minute. She had regular follow-up until the age of 6 months. At that age, her weight was 6700 g and her physical examination was normal. She was lost of follow-up since.

Figure 2. Post-pericardiocentesis chest radiography showing resolution of cardiomegaly with normalization of cardiac silhouette.

Discussion

Pericardial effusion (PE) is rare in neonates. Its clinical presentation depends on the speed and amount of fluid accumulation.² This can lead to tamponade and even death. The most common cause is iatrogenic because of central venous catheters.³ In a meta-analysis of 21 studies and 99 cases, pooled incidence of PE in neonates who had central venous catheters was 3.8‰. Mortality rate was 2.7%.¹

In a national American study of all cases of pericardial effusion, the lowest incidence was observed in neonates (0.04%).⁴ In a literature review, 34 patients were enrolled. The most common causes were central venous catheters (n = 21), Down syndrome (n = 3), and infections (n = 3). The pathogen agent was isolated in pericardial fluid in only one cas (candida albicans). The second case was caused by Escherichia coli isolated in blood culture. In the third case, PE was secondary to parainfluenza virus 3 which was isolated in nasopharyngeal swab.³

Pericardial effusion due to an E. coli infection is rare. To date, only case reports have been published. In 1979, Wynn described purulent pericarditis in a 64 hours aged neonate. Autopsy confirmed the diagnosis of pericardial effusion, and E coli was isolated from the blood culture.⁵ In 2006, Benjamin described pericarditis in a 10 days old boy. E. coli was isolated from blood samples obtained in the emergency room before referral. He was managed with pericardectomy, pericardial drainage, cefotaxime, and indomethacin. The intraoperative samples were negative.⁶

Even in E. coli sepsis, pericardial effusion remains rare. Lai described the epidemiology of invasive E. coli infection in 94 Chinese neonates, including early and late onset disease. No case of pericardial effusion was described.⁷

In our case, initial suspicion of a nosocomial infection was raised given the patient’s recent hospitalization. The isolation of E. coli with multidrug sensitivity underscores the potential for multidrug-resistant strains in neonatal units, especially in settings where broaded-spectrum antibiotics such as carbapenems and aminoglycosides are frequently used empirically. This highlights the importance of strict infection prevention measures, rational antibiotic stewardship and early targeted therapy guided by culture results.

The management of PE is variable, from surveillance in small asymptomatic effusions to pericardiocentesis, pericardectomy, and pericardial drainage.^3,8 In an American retrospective cohort, pericardial drainage was performed in 8.5% of neonates.⁴

Conclusion

Neonatal pericardial effusion due to Escherichia coli is an exceedingly rare but potentially life-threatening condition. Early recognition and timely pericardicentesis can be lifesaving.

Recommendations:

Use of AI tools: The authors did not use AI technology in the writing process.

Ethics and consent

Approval of the local ethics committee of Bechir Hamza Hospital was obtained (n°13/2022). Written informed consent was obtained from the patient’s guardian for the publication of clinical details.