Abstract 3119: Vascular Remodelling of the IMA-Graft Early and Late after Bypass Surgery
Objective: Flow mismatch between the supplying vessel and the perfusion region has been observed in patients with internal mammary artery (IMA) grafts. However, arterial graft remodelling has been postulated after CABG. The purpose of this study was to assess flow changes of arterial (IMA) and venous bypass grafts (SVG) after CABG.
Methods: 42 patients undergoing elective bypass surgery were included in the present analysis. Patients were studied intra-, early (4 months; n=12) and late (12 months; n=5) post-operatively. Coronary flow was measured intra-operatively with the transit time Doppler technique (CardioMed, Norway) in the IMA and one SVG. Post-operatively patients were re-catheterized 4 and 12 months after surgery and coronary bypass flow as well as flow reserve (CFR) was measured with the Doppler flow-wire technique in the IMA and SVG. Quantitative angiography was used to determine vessel size for calculation of absolute coronary flow. Wilcoxon Signed Ranks Test was used for paired observations and Mann Whitney U test for unpaired comparisons.
Results: Intra-operatively IMA flow was significantly lower than SVG flow (32±18 vs 56±30 ml/min; p<0.05). At 4 months flow in the IMA increased significantly to 45±21 ml/min; p<0.02 vs intra-operatively). At 12 months the IMA flow was 41±24 (ns vs 4 months). However, SVG flow remained unchanged over time and was 50±19 ml/min at 4 months (ns vs intra-operatively). Coronary flow reserve was abnormally low intra-operatively in the IMA (1.3±0.3) and SVG (1.6±0.6), but increased significantly to normal values in both grafts at early follow-up (2.3±0.2 for IMA and 2.5±0.5 for SVG; p< 0.05 vs intra-operatively), and increased further to 2.9±0.8 in the IMA graft at 12 months.
Conclusions: Bypass flow of the IMA is significantly reduced intra-operatively when compared to SVG. However, 4 and 12 months after CABG flow of the IMA is significantly increased and similar to SVG flow. This finding can be explained by an early flow mis-match of the native IMA and the large LAD perfusion territory. There is a vascular remodelling of the IMA graft with normalisation of coronary flow reserve over time.