There have been several studies of the effect of exercise on the htmodynamics in patients with cardiac septal defects and left-to-right shunts (Jonsson, Linderholm, and Pinardi, 1957;Scebat et al., 1957;Bruce and John, 1957;Swan, Marshall, and Wood, 1958;Stephens, Shafter, and Bliss, 1964), but some uncertainty remains about the conclusions. Since any attempt to understand the disability of such patients depends on these, we have attempted a further investigation into the changes occurring during exercise. This paper deals with the hemodynamics, and a following communication is concerned with ventilatory and metabolic aspects (Gazetopoulos and Davies, 1966).
SUBJECTS AND METHODSStudies were carried out on 34 patients, 18 with atrial septal defect (ASD), 12 with ventricular septal defect (VSD), and 4 with patent ductus arteriosus (PDA).The hemodynamic studies were performed during cardiac catheterization in the supine position. Routine data having been first obtained, a catheter was left in the pulmonary artery and an indwelling needle in a brachial artery. Spirometry was carried out and blood samples were taken; the catheter was then withdrawn to the lateral wall of the right atrium and further blood samples were obtained. In patients with ASD a second catheter, placed in the inferior vena cava, was withdrawn to the superior vena cava for further sampling. Paired samples were thus obtained from brachial artery, pulmonary artery, right atrium, or cavae for flow determinations by the Fick principle. Dye-dilution curves were also inscribed, as described below.The catheters were then replaced in their original sites and the patient exercised with either a springloaded leg exerciser (earlier studies) or a bicycle ergometer (Elema-Schonander-later studies). Pulmonary and brachial arterial pressures were continuously monitored and, after the fifth minute of exercise, Received July 21, 1965. * Present address: Alex. Soutson 8, Athens 134, Greece. 579 spirometry was repeated with the same sampling sequence as described above. Dye-dilution curves were also recorded during exercise. The spirometer was of the closed-circuit type (Donald and Christie, 1949) and oxygen consumption was corrected to STPD. Blood oxygen saturations were measured spectrophotometrically (Unicam model 600).As representative of mixed venous saturation we have used the mean of right atrial samples where the shunt was distal to this level, while in ASD the mean of two inferior caval and one superior caval value was taken. Dye-dilution curves were recorded on a Mark II Cambridge dye recorder with a high-resistance input circuit giving a linear response to concentration of dye (Gabe and Shillingford, 1961), Coomassie blue being used as indicator. Special attention was paid to injecting the same quantity of dye at the same site just above the pulmonary valve at rest and on exercise. The curves were replotted and extrapolated on semi-logarithmic paper, and the shunt calculated as a percentage of the pulmonary flow by planimetry. In cases with large leftto-r...