A 37-year-old man with tricuspid atresia, in whom a chronic right apical aspergilloma had stimulated formation of extensive aortopulmonary collateral circulation, suffered reversal of flow within his long-standing fistulous Glenn anastomosis. By reversing the flow through the pulmonary arteriovenous fistula and raising oxygen saturation in the right atrium, the acquired aortopulmonary collateral circulation prevented the increase in cyanosis which usually occurs when fistulous changes develop late after the Glenn operation.

A clinical study on the outcomes of Glenn anastomoses performed since 1987 in eight consecutive patients aged ≥16 years, and in two performed earlier, showed poor results. One badly selected patient died early as a consequence of high venous pressure, while a further seven had early complications. Seven of eight hospital survivors were followed for 1–10 (median 4.2) years with two deaths (1 and 4 years later). Of the remaining five patients, two improved temporarily, but increased arterial oxygen saturation was not maintained after 6 months. The two patients who had undergone a Glenn anastomosis 10 and 34 years earlier were shown to have pulmonary arteriovenous fistulas. The Glenn anastomosis in these older patients is associated with high rates of complication and appears not to give adequate palliation, particularly when it is the only source of pulmonary blood supply. In the adult, the Glenn anastomosis can be used as a staging procedure for Fontan-type surgery, but must be combined with another source of pulmonary arterial supply. Any adult having a Glenn anastomosis, particularly without another source of pulmonary arterial supply, should be warned of the possibility of worsening of cyanosis and symptoms. The second stage of the procedure may need to be performed soon after the first should the hypoxia prove intolerable.

We evaluated the concentrations of the brain and atrial natriuretic peptides in the plasma as markers of ventricular function and volume load in children with functionally univentricular hearts. We studied 7 children aged from 0.5 to 0.7 years with functionally univentricular hearts who had undergone a first palliative operation, and 10 children aged from 1.8 to 3.7 years who had undergone a bidirectional Glenn anastomosis at ages ranging from 0.4 to 1.0 year. As a control group, we studied 14 children without heart defects aged from 0.1 to 4.5 years. Levels of the brain natriuretic peptide were measured at 8.3 to 122 ng/l, with a mean of 52.8 ng/l, after the first palliative operation, compared to 0 to 16 ng/l, with a mean of 7.3 ng/l, after a bidirectional Glenn anastomosis, and 0 to 13.8 ng/l, with a mean of 5.9 ng/l, in the children serving as controls. Corresponding values for atrial natriuretic peptide were 17 to 203 ng/l, with a mean of 103 ng/l, after the first palliative operation, compared to 16 to 54 ng/l, with a mean of 29 ng/l, after the bidirectional Glenn anastomosis, and 12 to 52 ng/l, with a mean of 32 ng/l in the controls. Echocardiography showed that all the children with functionally univentricular hearts had normal ventricular function. Blood presssure, pulmonary arterial pressure, and arterial saturations of oxygen did not differ between the groups. We conclude, that in children with functionally univentricular hearts, the volume overload imposed on the heart after the first palliative operation is associated with increased production of brain and atrial natriuretic peptides, while after ventricular unloading, levels of the natriuretic peptides return to control values.