Left internal mammary grafting of the left anterior descending and diagonal arteries: considerations for grafting arrangement

K. Chaudhuri¹, Z. Andrew Jessop. Pullan², N. Smith^3
1Auckland City Hospital, Auckland, Auckland ²Greenlane Cardiothoracic Surgical Unit, Auckland City Hospital, Auckland, Auckland, Auckland ³Auckland Bioengineering Institute, Auckland, Auckland

Purpose: Uneven distribution of coronary stenoses can lead to flow competition in coronary bypass surgery involving sequential grafts. The objective was to determine the influence of differing stenoses severity when utilising sequential left internal mammary (LIMA) grafts according to the first touchdown being either the left anterior descending (LAD) or diagonal.

Methods: Using predictive computational fluid dynamics, a total of 300 patients with severe coronary disease were split evenly into two groups comparing a sequential LIMA to diagonal to LAD (‘diagonal-first’ configuration, n = 150) to a sequential LIMA to LAD to diagonal (‘LAD-first’ configuration, n = 150). Patients were further stratified into subgroups according to differing degrees of percent diameter stenoses (75 to 99%) between the LAD and diagonal territories. Transit time flowmetry (TTFM) parameters; mean graft flow (MGF); pulsatility index (PI); diastolic filling percentage (DF); backward flow percentage (BF) were calculated. Grafts were classified as unsatisfactory if MGF < 15 and PI > 5. Statistical analysis compared grafts using paired t-tests for TTFM parameters and chi-squared tests for proportion of unsatisfactory grafts. A p-value of < 0.05 was used for significance. Further sub-group analysis was conducted according to distribution of stenoses.

Results: TTFM measurements were influenced by relative degree of stenosis severity and grafting order. When the diagonal stenoses was considerably greater than the LAD, there was significantly more flow in the LIMA to diagonal graft for the ‘diagonal-first’ configuration (35.6 ± 11.3 vs 14.1 ± 7.6. p-value < 0.05). When this relationship of stenoses was reversed, and the LAD was considerably tighter than the diagonal, the LIMA to LAD graft MGF was similarly higher in the LAD-first group (57.8 ± 14.2 vs 39.2 ± 11.8. p-value < 0.05). The MGF remained higher in the respective first touchdown groups even when the stenoses difference was less (Table 1). The order in which the severity of stenoses was grafted determined the number of unsatisfactory grafts between configurations. When the terminal target was a less tight stenosis than the proximal target, there was a higher proportion of unsatisfactory grafts compared to when the final coronary target was a more severe stenosis (50.8% (61/120) vs 36.6% (44/120), p-value < 0.05).

Conclusion: The first touchdown of sequential LIMA to LAD/diagonal graft should be to a target with a less tight stenosis to avoid steal of flow. Orientation of the grafting arrangement can be chosen based on this. With similar degrees of stenoses, patient-specific computational modelling can help predict unsatisfactory grafting arrangements.

Identify the source of the funding for this research project: No funding, n/a.