Primary percutaneous coronary intervention (PPCI) is increasingly seen as the treatment of choice in acute myocardial infarction (AMI). Compared with thrombolysis, with PPCI recanalisation of the vessel is almost guaranteed, and, with elimination of the culprit epicardial stenosis, the risk of reocclusion is greatly reduced. There is, however, one advantage to thrombolysis: if it successfully restores epicardial flow, it can penetrate distal to the lesion and lyse further small microthrombi in the downstream myocardial bed. The effects of thrombus embolisation continue to be a problem in patients undergoing PPCI, in whom angiographic evidence of embolisation occurs in up to 15%,1 and are clearly related to adverse clinical and functional outcome.2 3 Once present, angiographic no-reflow, the end result of distal embolisation, is difficult to treat and disheartening for the operator.

In this issue of the journal, Reho and colleagues describe the successful use of a filter embolic protection device to retrieve clot during PPCI in a patient with high thrombus burden (see image on page 10.1136/hrt.2007.123273).4 They do not, however, simply leave the device distal to the thrombus during PCI. Instead, they pull back the open filter across the lesion and use the filter to “fish out” the thrombus, before recapturing the device proximal to the lesion. To our knowledge, this is the first description of such deliberate off-label use of this device and we applaud their bravery! However, such an approach is potentially loaded with iatrogenic complications. Firstly, assuming the device is appropriately sized to the vessel, the nitinol ring of the filter will be fully apposed to the vessel wall. Pulling back the expanded device runs the serious risk of endothelial denudation (with exposure of the intima and enhanced platelet activation) as well as causing severe vasospasm. Furthermore, it may remove not only thrombus, but also soft lipid-rich plaque, and such “atherectomy” with a filter device may seriously damage the vessel wall. It would have been interesting to microscopically assess the composition of the plaque.

Options to remove intracoronary thrombus include distal filter baskets, proximal or distal occlusion balloons with accompanying aspiration devices, and specifically designed thrombectomy catheters.5 Such myocardial protection devices were first studied during saphenous vein graft intervention, where their benefit in reducing distal embolisation and improving outcome is well-recognised.5 However, subsequent application of these devices to the setting of AMI intervention has been disappointing. The EMERALD trial randomised 500 patients to angioplasty with or without GuardWire distal protection.3 Use of myocardial protection conferred no benefit on the rapidity of ST-segment resolution, final infarct size or the incidence of 6-month major adverse cardiac events (MACE). This was despite the removal of visible debris in 73% of patients in the treatment arm. The lack of benefit was confirmed in the PROMISE trial, which randomised 200 patients undergoing PPCI to protection with or without adjunctive FilterWire.6

The thrombus aspiration catheters have also been evaluated, but generally in smaller trials. Napodano7 randomised 92 AMI patients with angiographic thrombus burden to PCI with or without thrombus aspiration using the X-Sizer. Although no benefit on 30-day MACE or LVEF was observed, use of the device significantly improved angiographic characteristics of post-procedure flow and myocardial blush grade. Had the study been more adequately powered, this might have translated into a clinical benefit. Several subsequent small randomised trials of thrombus aspiration support the beneficial effects on angiographic and electrocardiographic end points.8^–10 More recently, De Luca and colleagues reported that manual aspiration thrombectomy with the Diver catheter in 78 patients with STEMI and clear angiographic evidence of intracoronary thrombus significantly improved left ventricular remodelling, without improvement in clinical outcome.11 The TOPIT trial demonstrated a significant reduction in in-hospital MACE when thrombectomy was applied to thrombus-rich vessels, but, since non-native vessels were also included, extrapolation of the benefits to native vessels alone is difficult.12 Furthermore, use of the AngioJet in PPCI in the AIMI study paradoxically increased infarct size.13

Anecdotal case reports documenting the successful use of myocardial protection in patients with high thrombus burden undergoing PPCI have led to the increasing adoption of such a strategy. This conflict with the results of larger trials may be explained as follows. In case reports, there is always a very high thrombus burden, yet large-scale trials have used less stringent criteria in selecting AMI patients for myocardial protection. Furthermore, the true benefit of thrombus removal can only be assessed by quantitating volume of myocardium salvaged, as a function of myocardium at risk. Few studies have used SPECT to assess this end point. Importantly, the two studies which evaluated the effect of thrombus aspiration on myocardial salvage using SPECT failed to demonstrate a benefit.13 14 A recent meta-analysis suggests that the use of adjunctive mechanical devices to prevent distal embolisation is associated with better myocardial perfusion and less distal embolisation, but without an apparent improvement in survival.15

Routine use of myocardial protection devices in patients undergoing primary PCI cannot be advocated. In patients with a very high thrombus burden, however, reducing the debris prior to stenting would seem desirable and likely to achieve a favourable outcome. How best to achieve this remains unclear. There is strong evidence for the early upstream use of glycoprotein IIb/IIIa inhibitors and loading with 600 mg of clopidogrel. In terms of devices, simplicity is probably the key to success. We would recommend the use of a dedicated thrombus aspiration catheter for this purpose. There may be a role for adjunctive embolic protection systems, in particular to protect adjacent large vessels when the thrombus is near the ostium of the infarct-related artery.

Removal of visible debris and improved angiographic appearances will certainly make the cardiologist feel better. That both filters and thrombus aspiration devices improve angiographic characteristics in patients with high thrombus burden is not in doubt. Whether they improve the outcome for the patient remains questionable. A study to evaluate the complementary roles of aggressive thrombus aspiration (with vacuum) with downstream myocardial protection in patients with high-volume thrombus burden is required. Meanwhile, the risks and the benefits of adopting a thrombus removal technique should be assessed on a case-by-case basis.