J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.55 on 1 January 1997. Downloaded from BRIEF REPORTS Atraumatic Pneumopericardium in a Full-Term Newborn with Fetal

CPT Niel A.Johnson, Me, USA, and CPTJohnathan R. Gore, III, MC, USA

Pneumopericardium is an uncommon complica­ served. Apgar scores were 8 and 8 at 1 and 5 min­ tion of the neonatal transition period, occurring utes, respectively. in 1.3 percent of premature neonates in intensive The infant was taken to the neonatal intensive care settings. Its occurrence is usually associated care unit in stable condition for observation and a with some type of barotrauma or respiratory in­ routine sepsis workup for possible chorioam­ tervention, such as endotracheal intubation or ad­ nionitis. His temperature, blood pressure, heart ministration of continuous positive airway pres­ rate were normal, and his oxygen saturation was sure, and is almost exclusively a problem for the greater than 99 percent. His lungs were clear, and premature infant. there was no or ballooning of the chest The purpose of this case report is to describe or soft tissues of the neck. Plain chest radio­ the occurrence of pneumopericardium in a full­ graphs, done as part of this sepsis workup, showed term infant who had no apparent barotrauma and a pneumopericardium (Figures 1 and 2). The at­ whose symptoms were tachycardia and mild res­ tending neonatologist recommended intubation, piratory difficulty. We will also review the diag­ sedation, and examination by the pediatric tho­ nosis and current therapy of this potentially seri­ racic surgeon. ous complication in a newborn who develops After these initial interventions and consulta­ respiratory distress during transition. tion, several treatment options were considered. Conservative nonsurgical management of the Case Report pneumopericardium was selected because no evi­ A 3 770-g boy was born at term by normal sponta­ dence of tamponade was seen clinically or by neous vaginal delivery to a healthy 21-year-old echocardiogram. The pericardial air gradually re­ primigravida woman whose prenatal course was solved (Figure 3), and the infant was extubated on http://www.jabfm.org/ uncomplicated. Family physicians were in atten­ the 2nd day after delivery (Figure 4). The re­ dance at this delivery, which occurred at a large mainder of his hospital course was notable for a teaching hospital. Labor had progressed unevent­ blood culture that grew gram-positive cocci, for fully for approximately 8 hours while the mother which he was prescribed antibiotic therapy. The was receiving epidural anesthesia. One hour be­ source of this bacteremia was felt to be nosoco­

fore delivery, however, the mother's temperature mial, although was included in on 26 September 2021 by guest. Protected copyright. rose to 38.9°C, and the fetal heart rate increased the differential diagnosis. His recovery was un­ to more than 200 beats per minute. Antibiotics complicated, and he did well after discharge. were administered, and labor was augmented with intravenous oxytocin until delivery, which Discussion proceeded without difficulty over a small midline The cause of this infant's pneumopericardium episiotomy. The infant's tachycardia improved to was believed to be a small rupture of the pericar­ less than 180 beats per minute soon after delivery, dial sac adjacent to a nearby bronchial structure at but grunting and subcostal retractions were ob- the time of his initial breath. It was hypothesized that in his case, a congenital conduit allowed air to enter the during the first few Submitted, revised, 9 October 1996. breaths-a timewhen this conduit might have From the Department of Family and Community Medicine, been the path of least resistance. Increased heart Dwight D. Eisenhower Army Medical Center, Ft. Gordon, Georgia. Address reprint request to CPT Niel A. Johnson, MC, rate and breathing effort at the time of delivery as 723 Low Meadow Dr, Evans, GA 30809. a result of chorioamnionitis could have precipi-

Atraumatic Pneumopericardium in Newborn 55 J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.55 on 1 January 1997. Downloaded from

Figure 1. Initial anteroposterior chest radiograph Figure 2. Initial chest radiograph showing pericardial showing the classic pneumopericardial halo sign. air anteriorly. http://www.jabfm.org/ on 26 September 2021 by guest. Protected copyright. Figure 3. Chest radiograph, day 1, showing the slowly Figure 4. Chest radiograph, day 2, showing resolution resolving pneumopericardium. of the pericardial air. tated this event. As airway pressures normalized, fractured sternum or other related bony structure no further air passed into the pericardium, and was also considered but not supported radi­ the trapped air was absorbed. ographically or clinically. The actual incidence of such congenital anom­ Family physicians and other providers of alies is not known, but autopsy records show that neonatal care are often faced with the infant who they do exist. It is possible that physical trauma shows signs of respiratory distress. In full-term could have occurred during the birth process babies, in either the absence or presence of baro­ from the natural compression of the chest trauma, this respiratory difficulty could be due to through the pelvic outlet, but the incidence of delayed amniotic fluid absorption, meconium as­ such a complication should be much higher given piration, narcotic-induced depression, effects of the number of full-term infants born every day. A magnesium sulfate, and other related conditions.

56 JABFP Jan.-Feb.1997 Vol. 10 No. 1 J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.55 on 1 January 1997. Downloaded from

In premature babies, the list of potential causes neonatal intensive care settings. I ,2 Glenski and also includes hyaline membrane disease, sepsis, HalI2 reviewed 127 cases of neonatal pneu­ congenital malformations (eg, tracheoesophageal mopericardium from 1975 to 1984, and noted fistula), and fetal lung immaturity. The more that 121 occurred in neonates receiving positive­ common complications that can lead to pneu­ pressure ventilation; of the remaining 6 cases, 3 mopericardium include endotracheal intubation; infants were receiving negative-pressure ventila­ fracture of the rib, clavicle, or sternum; and a con­ tion, and 3 were not receiving any ventilatory genital heart defect resulting in a weak or incom­ support. The authors concluded that pneu­ plete pericardium, among others. Although most mopericardium is related to barotrauma, specifi­ complications would typically arise in the neona­ cally high peak inspiratory pressure, prolonged tal intensive care setting, familiarization with inspiratory time, and an elevated mean airway these complications better prepares the birth at­ pressure of 17 ± 10 cm of air. tendant to recognize them at the time of delivery. We found few cases of pneumopericardium The role of the family physician in this case caused by nontraumatic medical conditions. highlights an important teaching point. Because Bjorklund et aP reported pneumopericardium in the delivery occurred at a tertiary medical center, an infant born by forceps delivery without appar­ such resources as a neonatal intensive care unit, ent barotrauma. No other similar case of atrau­ in-house neonatology backup, and on-call pedi­ matic neonatal pneumopericardium appeared in atric thoracic surgery staff were readily available. our literature search. These resouces would not be available for many Pneumopericardium carries an overall mortal­ family physicians providing obstetric care at out­ ity of 70 to 75 percent, which is mostly attributed lying facilities. to coexisting illness, such as prematurity and res­ In retrospect, the management might have been piratory distress syndrome.3-s The mortality from overly aggressive, for the intubation itself, along pneumopericardium in full-term infants is not with gentle positive-pressure ventilation, placed known, but it is likely much less. In older children the baby at risk for further pneumopericardium and adults, atraumatic pneumopericardium can and other barotrauma, such as and be caused by achalasia, gas-producing infection, . At the time, however, the severe cough, chlorine gas exposure, cocaine in­ clinical picture of a tachycardic infant with a posi­ halation, emesis, barotrauma, parturition, and tive halo sign on his chest radiograph prompted a contact sports participation.6 quick decision for more secure, albeit invasive, Diagnosis of pneumopericardium, also called http://www.jabfm.org/ management. Because the neonatologist was pericardial emphysema, can be made by clinical much more accustomed to caring for premature and radiographic findings. In practice, however, infants that develop this problem, his judgment the diagnosis is almost exclusively made initially was understandably influenced by a population by plain film radiography, 1,6 where the finding of frequently known to require invasive measures, as air lining the pericardium, the classic halo sign,3 opposed to full-term infants who might require is diagnostic.7 Acute retrosternal pain radiating to on 26 September 2021 by guest. Protected copyright. only less-invasive means. Either way, the infant the back, neck, or shoulders is a common com­ did well, and the lesson for family physicians is plaint in adults. Cyanosis, dyspnea, and the classic that neonatal complications do occur, usually un­ ballooning of the soft tissues of the neck can be expectedly, and there are many ways to approach a observed,4 which when accompanied by the specific situation. The availability of resources and Hamman sign (mediastinal crepitation), suggest a consultants and the proximity to tertiary care are coexisting pneumomediastinum.6 Percussion of important variables in arriving at a treatment plan. the precordium can reveal tympany and, on aus­ We found several cases of traumatic pneumo­ cultation, dullness. Coexisting injury by pneumo­ pericardium on a MEDLINE review searching or pneumothorax is not uncommon back to 1966. Once considered rare, the number and is likely to complicate the accurate diagnosis of reported cases of traumatic pneumopericar­ of air in the pericardial sac.4 dium have increased from seven in 1970 to sever­ A chest radiograph will show the characteris­ alfold more in recent years, probably as a result of tic signs of pericardial air and is considered the di­ the increased use of high-pressure ventilation in agnostic study of choice. Transthoracic echocar-

Atraumatic Pneumopericardium in Newborn 57 J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.55 on 1 January 1997. Downloaded from

diography is sometimes indicated to look for evi­ Conclusion dence of . Electrocardiographic Pneumopericardium is an uncommon, but seri­ findings are not helpful, although older children ous, life-threatening complication of newborn and adults might show tachycardia and signs of life that usually results from barotrauma, but it .f> Cabatu and BrownH describe should be considered in a full-term newborn who the use of thoracic transillumination as a potential shows respiratory difficulty yet is born under ap­ diagnostic tool for this neonatal emergency. parently atraumatic conditions. The initial signs Management of the acute pneumopericardium can be unreliable in neonates, and chest radiogra­ is straightforward. Respiratory distress is man­ phy might be the only way to make an early diag­ aged by mechanical ventilation, if necessary, to nosis. Family physicians who deliver babies maintain adequate oxygenation. Cardiac tampon­ should be comfortable with not only recognizing ade is managed by needle pericardiocentesis to al­ the early signs of birth complications, such as low the heart chambers to fill properly. The un­ pneumopericardium, but also managing them. derlying cause is then sought7 after initiating resuscitative measures, including oxygen supple­ References mentation and sedation. Sepsis, intrinsic pul­ I. Neal RC, Beck DE, Smith VC, Null DM. Neonatal monary disease, and trauma should be treated as pneumopericardium with high-frequency ventilation. appropriate. Parents can be referred to a pediatric Ann Thorac Surg 1989;47:274-7. geneticist when congenital malformations are di­ 2. GlenskiJA, Hall Rr. Neonatal pneumopericardium: agnosed or suspected. Surgery is indicated to pro­ analysis of ventilatory variables. Crit Care Med 1984; 12:439-42. vide relief of the tamponade for large air collec­ 3. Bjorklund L, Lindroth M, Malmgren N, Warner A. tions causing distress. Spontaneous pneumopericardium in an otherwise Observation only might be indicated for mild healthy full-term newborn. Acta Paediatr Scand pneumopericardium when there are no signs of 1990; 79:234-6. cardiac wall motion abnormality or clinical signs 4. Emery RW, Foker J, Thompson TR. Neonatal pneu­ of distress, as the body can spontaneously resolve mopericardium: a surgical emergency. Ann Thorac the air without specific therapy.3,6 Intubation and Surg 1984;37:128-32. sedation might be indicated, as in our case, to al­ 5. Fiser DE, Walker WM. Tension pneumoperi­ low the pericardium a chance to heal for a brief cardium in an int~lIlt. Chest 1992; 102: 1888-91. period of time. Recurrence is almost guaranteed 6. Bartal N, Zaarura S, Marvan H. Spontaneous pneu­ if the original injury was due to ventilatory inter­ mopericardium and pneumomediastinum. Ann Otol http://www.jabfm.org/ Rhinol LaryngoI1990;99:1005-6. vention and the infant is kept on the ventilator for 7. Schwartz MZ, Palder SB, Tyson KR, Marr Cc. a prolonged period of time.],7 The management Complications of premahlrity that may require surgi­ of pneumopericardium clearly depends on the age cal intervention. Arch Surg 1988; 123: 1135-8. of the patient, the suspected cause, the available 8. Cabatu EE, Brown EG. Thoracic transillumination: tertiary neonatal medical support, and the degree aid in the diagnosis and treatment of pneumoperi­ of clinically observed respiratory compromise. cardiulll. Pecliatrics 1979;64:9511-60. on 26 September 2021 by guest. Protected copyright.

58 JABFP Jan.-Feb. 1997 Vol. 10 No.1