TWO FACTS have stimulated our interest in tuberculous pericarditis. First, and most important, there are indications that this disease is becoming amenable to treatment and may also be seen in early stages more often because of therapy of tuberculosis elsewhere. Second, the literature ascribes to acid-fast bacilli the major role in the production of adhesive pericarditis. This has appeared unlikely to us in the past, as it has occurred so rarely in tuberculous children seen at this hospital. It is true that in one of our cases the diagnosis was missed clinically, but in a survey of cases with autopsy in the past 30 years, this was the only one in which the diagnosis was not made ante mortem.
There have been 136 autopsies done on tuberculous children in the past 30 years. Six instances of tuberculous pericarditis were found. In five others a pathological diagnosis of the disease was