Left atrial function in ischemic heart disease assessed by intravenous digital subtraction angiography.

To investigate changes in left atrial morphology and dimensions during the cardiac cycle, the atrium was visualized by intravenous digital subtraction angiography (DSA). The study subjects consisted of 22 male patients whose average age was 54.5 +/- 8.6 years. They had ischemic heart disease without mitral valve disease and were in sinus rhythm. They were 11 patients with old myocardial infarction (OMI group) and 11 who had chest pain without evidence of infarction (AP group). DSA was performed in the continuous mode. Contrast material (35 ml) was injected at a rate of 18 ml/sec via a catheter in the superior vena cava and subtraction images were obtained at a speed of 30 frames/sec in the right anterior oblique projection. The left atrial and left ventricular margins were traced manually, their areas were calculated, and fractional changes in area were analyzed. The left ventricular ejection fraction (LVEF) was calculated by densitometry. Cardiac catheterization was performed in 16 patients and the left ventricular end-diastolic pressure (LVEDP) and mean pulmonary arterial wedge pressure (PAWP) were measured. The entire left atrium was clearly imaged using DSA. Phase analysis of the time-area curves in the right anterior oblique projection revealed that the left atrial area was maximal during left ventricular end-systole (%LA1 = 100%), it decreased during early left ventricular diastole (%LA2), and then increased slightly again during mid-diastole (%LA3). After left atrial contraction, the minimum area was obtained (%LA4). The left atrium showed a two-stage decrease in the area due to passive emptying and active contraction during left ventricular diastole. Passive emptying (%LA1-%LA2) was significantly less in the OMI group than in the AP group (6.3 +/- 3.6 vs 13.3 +/- 4.8%, p < 0.01, respectively). In all 22 subjects, passive emptying correlated with LVEF (r = 0.70, p < 0.001) and LVEDP (r = -0.58, p < 0.05). There was no difference in active contraction (%LA3-%LA4) between the 2 groups (26.0 +/- 5.7% in the OMI group, 28.2 +/- 8.4% in the AP group), and it did not correlate with LVEF or LVEDP. The ratio of passive emptying to active contraction [(%LA1-%LA2)/(%LA3-%LA4)] correlated with LVEF (r = 0.63, p < 0.01). These findings suggested that impaired left ventricular diastolic function and a relative increase in atrial contraction were present in patients with a lower LVEF. The %LA4 correlated with LVEDP and PAWP (r = 0.65, r = 0.63, p < 0.01, respectively). In conclusion, DSA proved to be a useful method for investigating left atrial morphology and function.