Pericardial Effusion After Cardiac Surgery in Children and Effects of Aspirin for Prevention Marie J. Bkland, MDCM, Marc Paquet, MD, James E. Gibbons, MD, Christo I. Tchervenkov, MDCM, and Anthony R.C. Dobell, MDCM
Seventy-four children aged 0.3 to 21.4 years (median 4.0) were fottowed echocardiographiiity on days 4,7,14 and 28 (* 2 days) after cardiac surgery to evaluate the incidence of postoperative pericardtai effusion, to identify the patients at greatest risk of developing an effusion and to evaiuate the use of aspirin as prophylaxis against pericardiai effushm. Pericardiai effuston was graded retattve to the size of the aortic root from grade 0 (no effusion) to grade 5 (larger than the aortic root dimension). Patients were randemiy divided into 2 groups: group 1(32 patients) received aspirin 60 mglkg/day for 7 days starting on the third postoperative day; group 2 (42 patients) received no aspirin. Forty-eight patients (65%) developed an effusion durfng the study period, 3 required pericardiocentesfs and 1 died of tampenade. Ail patients with tamponade had a grade 4 effuston. Age or type of operation did not alter the cumutative incidence of significant effusion. No patient with a grade 0 effusion on the first echocardiogram developed a grade 4 or 5 effusion. Results in groups 1 and 2 were similar. Perkardiai effusions are common in the fhst month after cardiac surgery. Patients with no effusion in the immediate postoperative period appear to be at lesser risk of developing a grade 4 effusion in the ftrst month after operation. Finally, aspirin prophylaxis against postoperative pericardial effusfons did not significantly atter the outcome in this smal series of patients. (AmJCardioi 1990;65:1238-1241)
he incidence of pericardial effusion in the first month after cardiac surgery in the pediatric population has not beenstudied, and the children at greatest risk for developing a clinically important effusion have not beenidentified. In addition, the possibility of preventing postoperativeeffusion through the use of prophylactic medication has not been examined. We designedthe present study (1) to evaluate prospectively by serial echocardiography the incidence of pericardial effusion in the first month after surgery for congenital heart diseasein the pediatric age group; (2) to determine which patients are most likely to developa significant effusion; and (3) to establish whether the prophylactic useof aspirin in the early postoperativeperiod can prevent the accumulation or attenuate the size of pericardial effusion after cardiac surgery.
T
METHODS Study protocol: Children over 3 months of age un-
dergoing open-heart surgery at our hospital between January 1985 and March 1988 were eligible for the study. On the day before surgery, patients were randomized into 2 groups: patients in group I were prescribed a 7-day course of aspirin (60 mg/kg/day orally divided every 6 hours) starting on the third day postoperatively; patients in group II did not receive aspirin in the first week after surgery. If additional analgesia or fever control was required in either group, meperidine, morphine, acetominophen, or a combination of these was used. Antiinflammatory medication (other than aspirin in group I) was not given to any patient during the first postoperativeweek. Six exclusion criteria were applied: (1) patients who could not take oral medications 3 or more days after surgery; (2) patients with bleeding disorders or hemorrhagic lesions; (3) patients receiving antiinflammatory medication(s); (4) patients allergic to aspirin; (5) patients having undergone the Fontan procedure; and (6) patients in whom informed consent was not obtained before randomization. gchecardiographic analysis: M-mode and 2-dimensional echocardiogramswere performed on postoperative days 4,7,14 and 28 (& 2 days) using an ATL MK600 ultrasound machine (Advanced Technology Laboratories). Studies were supervisedby a cardiologist and From the Department of Pediatrics (Division of Cardiology) and the Department of Surgery (Division of Cardiovascular Surgery), The reviewed by MJB, who were unaware of the results of Montreal Children’s Hospital, McGill University, Montreal, Canada. patient randomization. Manuscript received November 29,1989; revisedmanuscript received Anterior and posterior pericardial effusions were and acceptedJanuary 11,199O. measured by M-mode at the tip of the mitral valve in Address for reprints: Marie J. Bhland, MDCM, Division of Cardiology, Rm. C-302, The Montreal Children’s Hospital, 2300 Tupper systoleand diastole in the parasternal long-axis or shortaxis views, or both. We usedvariable signal damping to Street, Montreal, Quebec,Canada H3H lP3. 1238
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 65
emphasizethe separation betweenepicardium and pericardium for measurementpurposes.’ Verification of all measurementswas carried out by 2-dimensional echocardiography in the same views.2 We used the largest single measurementobtained for grading purposes.The aortic root was also measuredon M-mode from leading edge to leading edge at the aortic valve closure. Effusions were graded from 0 to 5 relative to the size of the aortic root (Table I). Effusions measuring 12 mm were considered insignificant and graded as 0. Patient population: Of the 287 open-heart casesperformed during the study period at our institution, 90 patients were enrolled in the study. Sixteen of these were not included in the results becauseof failure to obtain all 4 echocardiogramsat the designatedtime intervals. One patient died of tamponadebefore obtaining the fourth scheduled echocardiogram but was included in the study results. The study population therefore comprised 74 patients who underwent a total of 295 echocardiographic studies. The children ranged in age from 3 months to 21.4 years with a median age of 4.0 years. Thirty-two percent of the children (24 patients) were under 2 years of age. There were 40 male and 34 female patients. The operations were performed by either of 2 surgeonsand consistedof palliative or corrective proceduresfor a variety of congenital heart diseases with a preponderanceof atria1 and ventricular septal defect closures (Table II). Twenty-four patients (32% of the total group) required a ventriculotomy as part of the surgical intervention. No patient required oral anticoagulants during the study period. Thirty-two patients were randomly assigned to group 1, and 42 patients to group 2. The mean patient age in both groups was similar (4.7 vs 5.7 years), as was the proportion of patients who had a ventriculotomy (28 vs 36%). The pericardium was loosely closedin the majority of cases,but was left open in 5 of 32 (16%) of group I patients and 8 of 42 (19%) of group II patients. Statistical analysis: Differences betweenproportions were studied by Fisher’s exact test. A p value
Seventy-four children aged 0.3 to 21.4 years (median 4.0) were fottowed echocardiographiiity on days 4,7,14 and 28 (* 2 days) after cardiac surgery to evaluate the incidence of postoperative pericardtai effusion, to identify the patients at greatest risk of developing an effusion and to evaiuate the use of aspirin as prophylaxis against pericardiai effushm. Pericardiai effuston was graded retattve to the size of the aortic root from grade 0 (no effusion) to grade 5 (larger than the aortic root dimension). Patients were randemiy divided into 2 groups: group 1(32 patients) received aspirin 60 mglkg/day for 7 days starting on the third postoperative day; group 2 (42 patients) received no aspirin. Forty-eight patients (65%) developed an effusion durfng the study period, 3 required pericardiocentesfs and 1 died of tampenade. Ail patients with tamponade had a grade 4 effuston. Age or type of operation did not alter the cumutative incidence of significant effusion. No patient with a grade 0 effusion on the first echocardiogram developed a grade 4 or 5 effusion. Results in groups 1 and 2 were similar. Perkardiai effusions are common in the fhst month after cardiac surgery. Patients with no effusion in the immediate postoperative period appear to be at lesser risk of developing a grade 4 effusion in the ftrst month after operation. Finally, aspirin prophylaxis against postoperative pericardial effusfons did not significantly atter the outcome in this smal series of patients. (AmJCardioi 1990;65:1238-1241)
he incidence of pericardial effusion in the first month after cardiac surgery in the pediatric population has not beenstudied, and the children at greatest risk for developing a clinically important effusion have not beenidentified. In addition, the possibility of preventing postoperativeeffusion through the use of prophylactic medication has not been examined. We designedthe present study (1) to evaluate prospectively by serial echocardiography the incidence of pericardial effusion in the first month after surgery for congenital heart diseasein the pediatric age group; (2) to determine which patients are most likely to developa significant effusion; and (3) to establish whether the prophylactic useof aspirin in the early postoperativeperiod can prevent the accumulation or attenuate the size of pericardial effusion after cardiac surgery.
T
METHODS Study protocol: Children over 3 months of age un-
dergoing open-heart surgery at our hospital between January 1985 and March 1988 were eligible for the study. On the day before surgery, patients were randomized into 2 groups: patients in group I were prescribed a 7-day course of aspirin (60 mg/kg/day orally divided every 6 hours) starting on the third day postoperatively; patients in group II did not receive aspirin in the first week after surgery. If additional analgesia or fever control was required in either group, meperidine, morphine, acetominophen, or a combination of these was used. Antiinflammatory medication (other than aspirin in group I) was not given to any patient during the first postoperativeweek. Six exclusion criteria were applied: (1) patients who could not take oral medications 3 or more days after surgery; (2) patients with bleeding disorders or hemorrhagic lesions; (3) patients receiving antiinflammatory medication(s); (4) patients allergic to aspirin; (5) patients having undergone the Fontan procedure; and (6) patients in whom informed consent was not obtained before randomization. gchecardiographic analysis: M-mode and 2-dimensional echocardiogramswere performed on postoperative days 4,7,14 and 28 (& 2 days) using an ATL MK600 ultrasound machine (Advanced Technology Laboratories). Studies were supervisedby a cardiologist and From the Department of Pediatrics (Division of Cardiology) and the Department of Surgery (Division of Cardiovascular Surgery), The reviewed by MJB, who were unaware of the results of Montreal Children’s Hospital, McGill University, Montreal, Canada. patient randomization. Manuscript received November 29,1989; revisedmanuscript received Anterior and posterior pericardial effusions were and acceptedJanuary 11,199O. measured by M-mode at the tip of the mitral valve in Address for reprints: Marie J. Bhland, MDCM, Division of Cardiology, Rm. C-302, The Montreal Children’s Hospital, 2300 Tupper systoleand diastole in the parasternal long-axis or shortaxis views, or both. We usedvariable signal damping to Street, Montreal, Quebec,Canada H3H lP3. 1238
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 65
emphasizethe separation betweenepicardium and pericardium for measurementpurposes.’ Verification of all measurementswas carried out by 2-dimensional echocardiography in the same views.2 We used the largest single measurementobtained for grading purposes.The aortic root was also measuredon M-mode from leading edge to leading edge at the aortic valve closure. Effusions were graded from 0 to 5 relative to the size of the aortic root (Table I). Effusions measuring 12 mm were considered insignificant and graded as 0. Patient population: Of the 287 open-heart casesperformed during the study period at our institution, 90 patients were enrolled in the study. Sixteen of these were not included in the results becauseof failure to obtain all 4 echocardiogramsat the designatedtime intervals. One patient died of tamponadebefore obtaining the fourth scheduled echocardiogram but was included in the study results. The study population therefore comprised 74 patients who underwent a total of 295 echocardiographic studies. The children ranged in age from 3 months to 21.4 years with a median age of 4.0 years. Thirty-two percent of the children (24 patients) were under 2 years of age. There were 40 male and 34 female patients. The operations were performed by either of 2 surgeonsand consistedof palliative or corrective proceduresfor a variety of congenital heart diseases with a preponderanceof atria1 and ventricular septal defect closures (Table II). Twenty-four patients (32% of the total group) required a ventriculotomy as part of the surgical intervention. No patient required oral anticoagulants during the study period. Thirty-two patients were randomly assigned to group 1, and 42 patients to group 2. The mean patient age in both groups was similar (4.7 vs 5.7 years), as was the proportion of patients who had a ventriculotomy (28 vs 36%). The pericardium was loosely closedin the majority of cases,but was left open in 5 of 32 (16%) of group I patients and 8 of 42 (19%) of group II patients. Statistical analysis: Differences betweenproportions were studied by Fisher’s exact test. A p value
