An investigation of transit-time flowmetry parameters in coronary artery by-pass grafting

Coronary artery disease (CAD) remains the leading cause of early death and illness in developed countries. Today, patients with CAD are treated either with medical therapy, percutaneous coronary intervention, or with coronary artery by-pass grafting (CABG). Although CABG remains the reference-standard for revascularisation of multi-vessel CAD, long-term patency of by-pass grafts continues to be a major challenge. Literature suggests that perioperative graft failure occurs in up to 11% of by-pass grafts, affecting roughly 10% of all CABG patients, and hence numerous interventions have over the years been investigated to prolong graft patency and to identify grafts that may eventually go on to fail. Potentially failing grafts should ideally be identified during the operation to enable graft revision, before the chest is closed. The conventional approach of visual inspection or palpation of the graft to assess flow can be deceiving, and European guidelines on myocardial revascularization now recommend intraoperative determination of graft flow with Transit time flowmetry (TTFM). Today, TTFM is the most commonly applied technique for intraoperative graft flow assessment. TTFM measures four flow parameters, mean graft flow (MGF), pulsatility index (PI), percentage of diastolic filling (%DF), and percentage of backward flow (%BF), but guidelines only provide cut-off values for MGF and PI, with no further distinction on how these parameters are affected in different CABG settings, and hence without a deeper understanding of the interrelation of measured parameters. The work comprising this dissertation has sought to evaluate the utility of the TTFM technique by assessing the accuracy and precision of the technique in measuring blood flow in addition to assessing all TTFM parameters in different CABG settings. In a retrospective observational study of 60 CABG patients, the accuracy and precision of the TTFM technique was evaluated by correlating measured TTFM blood flow to routinely determined free blood flow of the left internal mammary artery prior to grafting. In this study, it was found that (1) Overall overestimation was 7.1%, with a standard deviation of TTFM to true flows of 16.3%, with largest deviations from true flow seen with oversized TTFM probes. (2) TTFM probe oversizing was associated with poor acoustic coupling, reflecting poor probe to graft contact. (3) The study suggests that flow overestimation may be suspected with flows above 68 ml/min, but a true low flow is rarely deemed normal/too high. In three separate observational studies of a total of 268 patients, all TTFM parameters were evaluated according to surgical approach (on- vs. off-pump CABG), choice of conduit (arterial vs. venous conduits), and grafted coronary territory (by-pass grafts to the right or left sided coronary arteries, respectively). The results of these three studies were as follows: (1) compared to off-pump surgery, MGF is higher for arterial grafts (50±35 vs. 38±26 ml/min, p<0.001) and %BF is lower for saphenous vein grafts (1.8±6.2 vs. 2.6±6.2%, p=0.003) during on-pump surgery, (2) arterial grafts have significantly higher %DF (71±8 vs. 64±11%, p<0.001) and %BF (2.3±3.2 vs. 1.7±3.2%, p=0.002) than venous grafts, but comparable MGF and PI, and (3) flow parameters are comparable between left- and right-sided saphenous vein by-pass grafts. Finally, in a randomized clinical study of 35 patients, TTFM parameters of externally stented vein grafts were compared to non-stented vein grafts. In this study, it was demonstrated that external stenting of saphenous vein grafts does not affect intraoperative flow parameters. The conclusions of this body of work are that, with a slight overall overestimation, TTFM is a good indicator of blood flow with a clinically acceptable precision only if appropriate angulation and probe size is ensured prior to measurement. Overestimation may be expected with flows >68 ml/min, but most importantly in situations with oversized TTFM probes. Furthermore, external stenting and coronary territory does not significantly affect TTFM parameters of saphenous vein grafts. However, research aiming to delineate TTFM cut-off values for graft patency must take into account the choice of conduit (arterial or venous) as well as the surgical approach (on- or off-pump).