Postpericardiotomy syndrome: no evidence for a viral etiology

Background, Postpericardiotomy syndrome has been considered a disorder indu ced by viral infection. This conclusion is based on serologic criterions, b ut these may be unreliable following either cardiopulmonary bypass or trans fusion therapy. Previous studies have not verified the proposed etiology ei ther by isolation of viruses, or by detection of their genome. We sought, t herefore, to clarify the role, if any, of viruses in this syndrome. Methods and Results. We studied prospectively 149 children aged from 6 months to 1 6 years who were undergoing open heart surgery Blood samples were collected from all prior to operation, and again 7 to 10 days post-operatively, and 47 were sampled at the time of development of symptoms of pericardial invol vement. Serums were analyzed for the presence of IgM and IgG antibodies to cytomegalovirus, herpes simplex virus, and Epstein-Barr virus. The polymera se chain reaction was used fur amplification when assessing the genome of t he enteroviruses. Cultures for viruses were established on samples of stool , urine, and throat swabs collected 7 days post-operatively, and at the tim e of postpericardial symptoms. Pericardial fluid obtained from 5 patients w ith the syndrome was cultured for viruses, and tested for enterovirus genom e. On die basis of clinical and echocardiographic findings, 34 children wer e determined to have definite evidence of the syndrome, 13 were considered to have possible evidence, and the results from these patients were compare d to those from patients with no pericardial symptoms, the latter being mat ched fur age and transfusion status. We isolated viruses from one or more s ites in five patients with definite evidence (16%), from one (9%) of those with possible evidence, and from seven (19%) of the controls. All serums an d pericardial samples were negative for enterovirus genome. IgM antibodies were found in only 5 patients, three with symptoms of pericardial involveme nt and two without. Rates of seroconversion to IgG for the viruses were low er in the patients with symptoms of pericardial involvement compared to con trols, but were strongly influenced by transfusion status. Conclusion. Our study has provided no evidence to support a viral etiology for the postperi cardiotomy syndrome.