Direct stenting in ST-elevation myocardials infarction: convenient, but not improving outcomes

This editorial refers to ‘Clinical impact of direct stenting and interaction with thrombus aspiration in patients with ST-segment elevation myocardial infarction undergoing percutaneous coronary intervention: Thrombectomy Trialists Collaboration’†, by K.D. Mahmoud et al., on page 2472. In the present issue of the journal, the Thrombectomy Trialists Collaboration reports important new data on direct stenting in conjunction with aspiration thrombectomy during primary percutaneous coronary intervention (PPCI) in acute ST-elevation myocardial infarction (STEMI).1 The findings provide new mechanistic insight and may impact on interventional practice. In acute STEMI, PPCI is highly successful in re-opening the infarct-related artery and achieves Thrombolysis in Myocardial Infarction (TIMI) grade 3 flow in >90% of the patients. Nevertheless, reperfusion at the tissue level in the area at risk is frequently incomplete, resulting in suboptimal myocardial salvage and poor clinical outcome. Apart from ischaemic damage to the microvasculature and reperfusion injury induced by regional inflammatory responses,2 distal embolization of fragmented thrombus and atheromatous debris from the culprit lesion is considered as a major underlying mechanism for compromised myocardial reperfusion after PPCI (Take home figure).3 As angiographic evidence of distal embolization, closure of terminal branches has been reported in 9–15% of the patients undergoing PPCI. In one study, this was associated with an eight-fold increase in 5-year mortality. Autopsy studies4 and experience with distal filter devices capturing debris and thrombus material3,5,6 suggest that angiography underestimates the true extent of distal embolization considerably. Mechanisms of curtailed myocardial reperfusion after PPCI in STEMI and role of aspiration thrombectomy and direct stenting. Left inset: angiographic image of distal embolization; the black arrow shows thrombus on the culprit lesion and white arrows indicate occluded distal branches. Reproduced with permission from Henriques et al.3 Right inset: autopsy study after sudden death in acute coronary syndrome showing microvascular obstruction by platelet aggregate. Reproduced with permission from Davies et al.4 Mechanisms of curtailed myocardial reperfusion after PPCI in STEMI and role of aspiration thrombectomy and direct stenting. Left inset: angiographic image of distal embolization; the black arrow shows thrombus on the culprit lesion and white arrows indicate occluded distal branches. Reproduced with permission from Henriques et al.3 Right inset: autopsy study after sudden death in acute coronary syndrome showing microvascular obstruction by platelet aggregate. Reproduced with permission from Davies et al.4 These findings prompted considerable interest in the development of procedural device-based strategies to minimize distal embolization during PPCI. Among the most thoroughly studied approaches are distal protection devices, aspiration thrombectomy, and direct stenting. Three randomized studies including between 200 and 668 patients investigated filter devices for distal protection. Although visible debris could be captured in up to 73% of the filters retrieved, these studies failed to show a benefit in terms of reperfusion measures, final infarct size, or clinical outcome.5–7 Contrary to the disappointing findings with filter wires, the TAPAS (Thrombus Aspiration During Percutaneous Coronary Intervention in Acute Myocardial Infarction Study) trial randomly assigning 1071 patients to PPCI with or without aspiration thrombectomy demonstrated a significant survival benefit of aspiration thrombectomy at 1 year.8 Yet, two subsequent larger randomized studies, TASTE9 (Thrombus Aspiration in ST-Elevation Myocardial Infarction in Scandinavia) and TOTAL10 (Trial of Routine Aspiration Thrombectomy with PCI versus PCI Alone in Patients with STEMI), could not confirm this finding, and TOTAL even reported a significantly increased risk of stroke associated with aspiration thrombectomy. A study-level meta-analysis, including the three landmark trials as well as 15 smaller randomized trials, reviewed available evidence on PPCI with aspiration thrombectomy as compared with conventional PPCI.11 This revealed only trends towards decreased risk of major adverse cardiac events (MACE) and towards decreased all-cause mortality, but no conclusive evidence of a benefit of aspiration thrombectomy. It is the merit of the Thrombectomy Trialists Collaboration that they combined individual data of the 18 306 patients studied in TAPAS, TASTE, and TOTAL. Notably, this data set comprises 85% of the patients studied in randomized controlled trials on aspiration thrombectomy. The Thrombectomy Trialists Collaboration thus enabled a comprehensive individual patient data meta-analysis that provided considerably greater power to assess low-frequency events than study-level meta-analyses and also enabled meaningful subgroup analyses.12 As their central finding, the authors report a numerically lower risk of cardiovascular death at 30 days (2.4% vs. 2.9%) after aspiration thrombectomy as compared with conventional PPCI at the expense of a numerically higher risk stroke or transient ischaemic attack (0.8% vs. 0.5%), but none of these differences reached statistical significance (P = 0.06 for both comparisons). Both differences in outcome became statistically significant in the subset with high thrombus burden (TIMI thrombus grade ≥3), albeit with an insignificant P-value for interaction (Pint = 0.32 or 0.34, respectively). Thus, there was no proof of a clinical benefit of aspiration thrombectomy with either routine application or selective use for high thrombus burden. Direct stenting without balloon pre-dilatation is considered another means for prevention of distal embolization by minimizing manipulation of the lesion and fixing loose material to the vessel wall that otherwise might have been dislodged during lesion preparation.13 In the setting of myocardial infarction, there are only five small trials with 754 patients enrolled that compared direct stenting with conventional stenting, which have been summarized in a recent meta-analysis.14 As compared with conventional stenting, direct stenting appeared to improve reperfusion as evidenced by a significant improvement of ST-resolution and a significant reduction in no reflow, and was also associated with a significant reduction of in-hospital cardiac death. Nevertheless, considerable uncertainty remained as to the role of direct stenting in STEMI because of the small sample size of the trials in MI. Moreover, none of these trials involved contemporary drug-eluting stents and adjunctive medical treatment. The putative benefits of direct stenting impact on the role of aspiration thrombectomy, given that aspiration thrombectomy facilitates direct stenting. In line with this notion, it has been speculated that the differences in treatment effect on survival between TAPAS and TOTAL, or TASTE might have been due to a more systematic exploitation of the potential of aspiration thrombectomy to enable direct stenting. Based on these considerations, the Thrombectomy Trialists Collaboration built on their large data set to perform the largest observational study comparing direct stenting with conventional stenting during PPCI, thereby also addressing the interaction with aspiration thrombectomy.1 In this data set, 32% of the 17 329 patients included underwent direct stenting. Confirming the facilitation of direct stenting by aspiration thrombectomy, direct stenting was performed significantly more often in patients assigned to aspiration thrombectomy than in patients assigned to conventional PCI (41% vs. 22%; P < 0.001). Direct stenting saved 10 mL of contrast volume and shortened the fluoroscopy time by 2.2 min (P < 0.001 for both comparisons). Of note, this benefit was gained with a background of random allocation to aspiration thrombectomy. It is questionable whether a similar benefit would be gained if an otherwise unneeded aspiration thrombectomy is performed to enable direct stenting. To compensate for the non-randomized nature of the comparison, propensity matching was performed, resulting in an analysis cohort of 10 944 patients. Comparing direct stenting with conventional stenting, the analysis did not yield any evidence of benefit or harm of direct stenting. Moreover, there was no significant interaction with thrombus aspiration for any of the clinical outcome variables investigated. Specifically, 30-day incidences of cardiovascular death or cerebrovascular events were low and not significantly different between direct stenting and conventional stenting (1.7% vs. 1.9%, adjusted P = 0.60 for cardiovascular death; 0.6% vs. 0.4%, adjusted P=0.99 for stroke/transient ischaemic attack). Likewise, there were no significant differences in all-cause mortality, myocardial infarction, and stent thrombosis. The 1-year incidence of vessel revascularization was lower after direct stenting than after conventional stenting (4.3% vs. 5.6%). Yet, in the adjusted analysis, only a trend prevailed (adjusted P = 0.07). Nevertheless, this finding is important. It has been argued that direct stenting may result in inadequate sizing of the stent diameter or length, resulting in incomplete lesion coverage or late malapposition, which may increase the risk of stent thrombosis and restenosis. It is thus reassuring that the current analysis refutes these concerns by showing a trend in the opposite direction. In their current communication, the Thrombectomy Trialists Collaboration also gives some mechanistic insight by addressing measures of reperfusion, which were assessed in TAPAS and TOTAL. In 6534 propensity-matched patients of these trials, incomplete ST-segment resolution (<70 %) was less frequent after direct stenting than after conventional stenting (3 % vs. 35%) as was the proportion of patients with myocardial blush grade 0 or 1 (4.7% vs. 5.7%). Yet, a statistically significant difference could not be maintained after adjustment [adjusted odds ratio (OR) 0.84, 95% confidence interval (CI) 0.70–1.01 and OR 0.99, 95% CI 0.66–1.47, respectively]. Only a trend towards improved ST-segment resolution by direct stenting was noted (P = 0.06). Thus, the current analysis failed to prove a favourable effect of direct stenting on myocardial perfusion which had been the putative mechanism for a clinical benefit. There is no doubt that distal protection devices capture embolized material during PPCI and that aspiration thrombectomy removes thrombus and reduces thrombus burden; it is also highly intuitive that direct stenting reduces distal embolization. A positive effect of these measures on distal reperfusion, however, has been hard to demonstrate. Although some studies suggested such an effect, larger data sets yield equivocal results. The current large patient-level meta-analysis strengthens this point for direct stenting in conjunction with aspiration thrombectomy.1 It demonstrates that improvement of distal reperfusion, if at all present, will be limited. This limited, if not absent, effect on distal perfusion is conceivable considering the pathophysiology of reperfusion in STEMI (Take home figure). At the time of PPCI, the myocardial microvasculature is already altered by ischaemic injury and by prior distal embolization of thrombus and plaque material during evolving STEMI. Further damage may occur by reperfusion injury. Thus, device-based strategies to reduce distal embolization can only modify one element among many of curtailed reperfusion in STEMI. This limited effect may not suffice to translate into a meaningful clinical benefit in the current era of optimized PCI and adjunctive treatment. Given that the mortality in uncomplicated STEMI is already very low—1.8% at 30 days and 2.9% at 1 year in the current data set—it will be hard to show further improvement by any means. From a procedural standpoint, direct stenting in STEMI is convenient as it shortens the procedure and saves contrast volume. The current analysis demonstrates that direct stenting in STEMI can be performed without concerns about subsequent increased risks of stent thrombosis or restenosis due to inadequate stent sizing. Yet, apart from procedural aspects, there is no appreciable benefit of direct stenting, as shown by the current meta-analysis. This means that interventional cardiologists can choose the stenting approach purely based on procedural considerations. There is no need to perform aspiration thrombectomy to facilitate direct stenting. This is consistent with current guidelines that do not recommend routine use of aspiration thrombectomy.15 However, aspiration thrombectomy may be useful in specific settings, such as high thrombus burden. When used in this setting, interventional cardiologists should take appropriate measures to minimize the risk of systemic embolization such as deep intubation of the guiding catheter and maintenance of negative pressure on the thrombectomy device during retrieval. Conflict of interest: the University Heart Centre Freiburg - Bad Krozingen has received research grants and speaker honoraria from Biotronic, Medtronic, and BostonScientific.