Spontaneous and interventional coronary microembolisation
- Correspondence to:
Dr Raimund Erbel, Department of Cardiology, University Essen, Hufelandstrasser 55, D-45122 Essen, Germany;
- CFR, coronary flow reserve
- PCI, percutaneous coronary intervention
- PTCA, percutaneous transluminal coronary angioplasty
- TIMI, thrombolysis in myocardial infarction
During recent years new imaging techniques have significantly improved the option to study in vivo coronary macro- and microvascular morphology and perfusion, including intracoronary ultrasound and Doppler, positron emission computed tomography as well as magnetic resonance imaging and high frequency transthoracic echocardiography.
SPONTANEOUS CORONARY MICROEMBOLISATION
The development of coronary atherosclerosis can be subdivided into five stages according to the American Heart Association.1 Atheroma and fibroatheroma, stage IV and V lesions, are characterised by lipid core formation. When the ratio of the lipid core to total plaque size increases and the fibrous cap thickness decreases < 60 μm, the risk of rupture is enhanced particularly if an infiltration of macrophages indicates signs of inflammation.2 As these lesions are prone to rupture, they are regarded as vulnerable or plaques at risk.3 Plaques which are vulnerable are showing more vascular compensatory remodelling than non-vulnerable plaques.4 Clinical events occur when plaques ulcerate (VIa lesions), show intramural bleeding (VIb lesions), and form mural or occlusive thrombus based on erosion or plaque rupture (VIc lesions).5,6
Apparently healthy subjects undergoing necropsy have experienced plaque rupture in 9% of cases, and patients with diabetes and hypertension in 17%.7,8 Rupture ( fig 1) accounts for > 60% of all thrombi associated with sudden coronary death and acutemyocardial infarction, plaque erosion for 35%, and calcified noduli for 2–6 %.2 Signs of plaque rupture are found in 85% of patients with unstable angina, but also in 15–25% with stable angina.5 Pathological and clinical studies have revealed multiple lesions, not only limited to a single coronary vessel but also in more than one vessel in unstable angina in mean (SD) 1.3 (1.4) hearts and in stable angina in 1.1 (1.3) hearts.2 Clinically 2.1–2.6 complicated lesions per patient were found in acute coronary syndromes with elevated cardiac markers, with a wide range from 0–6 lesions.9,10 These lesions are not limited to the dominant vessel with the culprit lesion, but are also found in the remaining coronary tree.
The development of multiple layering suggests repetitive, periodically occurring plaque rupture during life. Meanwhile the healing process of plaque rupture itself could be detected by intravascular ultrasound. Thrombus formation within the ulcer seems to be part of this healing process which results later in an attachment of the fibrous cap to the vessel wall, leading to plaque thickening. This mechanism seems to lead to progression of coronary artery disease, particularly when the effect of multiple events at the same site is taken into account.
Quantitative pathological anatomic analysis showed in fibroatheroma, thin walled atheroma, and ruptured plaques a mean (SD) necrotic core area of 1.2 (2.2), 1.7 (1.1), and 2.8 (5.5) mm2 and a ratio of the necrotic core to the plaque size of 15 (20)%, 23 (17)%, and 34 (17)%, respectively.2 Using similar definitions by intravascular ultrasound quantification the plaque ulcer measured 12 (13) mm3 within a plaque size of 72 (53) mm3.11 The ulcer volume measured up to 50 mm3.11 When plaque rupture occurs, this amount of lipid core material (“gruel”) can be washed out and lead to distal embolisation of the coronary tree. Depending on the size of the emboli, macro- or microinfarcts develop. In addition, shed membrane microparticles, with procoagulant potential and apoptosis as a critical determinant of plaque thrombogenicity after plaque rupture, are found. Also the exposed tissue of the plaque ulcer is a strong promotor of coronary thrombosis and contains high concentrations of tissue factors. Thrombus formation within the ulcer can develop and can be washed out, again partly explaining the observation of signs of multiple events. Thrombus formation also seems to be the first step in the healing phase. It also prevents further embolisation of the lipid core material. The atheromatous material often represents the source of a larger thrombus apposition and may immediately or subsequently lead to an enlargement of the infarct zone.
Exposure of denuded intimal layers—plaque erosion—may also produce mural or occlusive thrombus formation and also, like plaque ulceration, lead to type VIc lesions which are clinically most often found in unstable angina and acute myocardial infarction. Luminal narrowing with increased flow velocity can induce disruption of mural thrombotic material from type VIc lesions and also result in microembolisation. Even in high grade coronary artery stenosis, cycle flow variation with continuous platelet aggregation formation, increases in flow velocity, thrombi disruption, and microembolisation have been described. Therefore it is not surprising to find signs of coronary microemboli in patients with stable and unstable angina. The major difference between these clinical situations seems to be the number of episodes which are present in unstable angina (about 85%) and stable angina (70%), with microemboli in nearly 50% and 40%, and microinfarcts in about 45% and 30% of cases, respectively. Recurrent embolisation caused by remodelling of the platelet thrombus is also found. Platelet-rich thrombemboli have been demonstrated in up to 80% of patients with unstable angina and sudden death.12
If microembolisation occurs, a wide variety of clinical pictures can be induced, ranging from asymptomatic episodes only detected following analysis of cardiac markers and ECG recordings, up to the no reflow phenomenon. Signs of subsequent microinfarcts are found even after exercise testing in coronary artery disease, depending on the extent of the disease and the need for further revascularisation. Vulnerable plaques are characterised by signs of inflammation, which can clinically be detected by determining cytokines, but also C reactive protein as well as fibrinogen. What is not yet understood is whether these signs are related to vulnerable plaques or are induced by microembolisation.
The microemboli induced symptoms seem unresponsive to glyceryl trinitrate, but the features have to be analysed in much more detail in the future. Signs of microinfarcts (infarctlets), based on analysis of cardiac markers and ECG recordings and elevation of C reactive protein, are much more often detected than previously reported. Many patients, previously classified as having unstable angina, are now defined as having non-ST segment myocardial infarction. Thus, the number of patients with myocardial infarction has increased.
Of utmost importance is the finding that plaque rupture often occurs in the absence of a flow limiting stenosis, and can even be present when coronary angiography is negative or suggestive of a normal anatomy. This is found in up to 15% of unstable patients. Nearly 90% of patients with acute myocardial infarction have stenosis < 70%.13
Since significant coronary luminal narrowing cannot explain the symptoms and signs of myocardial ischaemia, other pathophysiologic mechanisms have to be considered in addition to microembolisation:
intermittent coronary spasm enhanced by plaque rupture with activation of platelets and leucocytes and release of vasoactive mediators, such as serotonin and endothelin
imbalance of vasoconstriction and vasodilation caused by endothelial damage and dysfunction, which may be enhanced by plaque haemorrhage (type VI b lesion)
cyclic flow variations with recurrent platelet aggregation and thrombus formation and washout
thrombus formation after plaque fissuring or rupture (type VI c lesion) which is not totally blocking the coronary artery, but may or may not embolise, as demonstrated by angioscopy and intravascular ultrasound
plaque rupture and ulceration with microembolisation of plaque debris (type VI a lesions).
INTERVENTIONAL CORONARY MICROEMBOLISATION
When Andreas Grüntzig introduced percutaneous transluminal coronary angioplasty (PTCA) in 1977, he thought that embolisation may be induced, but he found no wash out of plaque material into the distal coronary artery in experimental studies. Using PTCA a rise of cardiac markers was noticed only rarely, despite the fact that ECG recordings detected ST-T segment changes in a larger group of patients. The systemic determination of the new cardiac markers troponin T or I and the introduction of coronary stenting, as well as rotational angioplasty in percutaneous coronary intervention (PCI), led to a 5- to 10-fold increase of cases with so called “infarctlets”.14,15 The prevalence of these laboratory findings was higher in patients with unstable than stable angina (34% v 10%), higher in patients after stenting than after PTCA (22% v 4%), and higher after recanalisation than after angioplasty of coronary stenosis. In patients undergoing coronary rotational angioplasty the highest percentage of patients with “infarctlets” (37%) was observed. Independent predictors were high grade complicated type C lesions (American College of Cardiology/American Heart Association classification) and coronary dissection induced by PCI. Also plaque burden, but not cross sectional area at the reference or lesion site, was significantly related to rise of cardiac markers after PCI.14,15
Analysis of coronary flow reserve (CFR) before and after PCI demonstrated that an improvement and even a normalisation could be reached.15 But an abnormal flow reserve persisted after PTCA in 35% of patients, with a concomitant reduction in relative coronary flow reserve in up to 80%. Using coronary stenting the rate of normalised CFR could be increased compared to PTCA alone. Even so, in more than 20% of patients a reduced flow reserve was present, despite an open vessel with no residual stenosis and normal flow characteristics in other areas of the heart.15 But an elevation of baseline coronary flow velocity could be detected, which was highest during rotational angioplasty and lower after stenting, resulting in a reduced CFR despite an increase in hyperaemic (maximal) flow. Experimentally, an increase in baseline coronary flow velocity was observed when coronary microembolisation was induced leading to a reduced CFR, because maximal flow velocity remained constant or decreased slightly.16 With further microembolisation baseline flow velocity decreased, too. These findings were dependent on the size of the microemboli.16 Clinically, coronary flow impairment was associated with an elevation of both troponin and creatine phosphate kinase, supporting the idea of procedural related embolisation caused by plaque disruption, squeezing, and redistribution, resulting in myocardial injury.14 The overall better correlation between the cardiac marker troponin T/I, outcome, and CFR is explained by the experimental finding that an increase in baseline flow velocity preferably characterises minor forms of embolic myocardial injury.14 The important role of inflammation in the presence of vulnerable plaques could be demonstrated by a close correlation between changes of coronary flow velocity after intervention and the pre-interventional level of C reactive protein, indicating that plaques with signs of inflammation are more prone to plaque rupture inducing larger showers of microemboli than those without inflammation. It may be, however, that our techniques are not yet sensitive enough. Catheters enabling thermography seem to be a very promising tool.
Postprocedural cardiac marker increase is neither rare nor a prognostically insignificant event, which has been shown by most studies.17 Importantly, minor rather than major myocardial injury can be found frequently, with troponin being of higher diagnostic value than creatine kinase. The correlation of cardiac mortality and overall major adverse cardiac events was found to be stronger with postprocedurally increased troponin than with postprocedurally increased creatine kinase.17 Patients at risk can be identified by determination of troponin within 24 hours after intervention.
Statins are highly effective in primary and secondary prevention of acute myocardial infarction and cardiac death. They induce a stabilisation of the coronary plaques by reducing the inflammatory response and the lipid core, and increasing the collagen content. The endothelial function also improves. These pleiotropic effects explain the cardioprotective profile of statins in ischaemia/reperfusion models. Pretreatment with statins resulted in a more than 90% reduction in the incidence of postprocedural elevation of cardiac markers and improvement of coronary flow, supporting a cardioprotective effect in coronary interventions.18 This effect can also be related to a reduction in circulating concentrations of C reactive protein with less severe postprocedural microvascular impairment, as a direct proin- flammatory effect of C reactive protein on endothelial cells. Thus, preprocedural statin treatment might improve postprocedural outcome by modulating the target lesion and/or the myocardial microcirculation.
Also preinterventional β blocker treatment was found to reduce the risk of peri-interventional myocardial injury. Glycoprotein IIb/IIIa receptor inhibitors have also been shown to prevent at least major forms of peri-interventional myocardial injury, but not smaller forms as a result of large amounts of embolising debris in saphenous venous bypass grafts.
In acute myocardial infarction reopening of the occluded coronary vessel by thrombolytic treatment has proved successful in improving ventricular function and/or survival. Combined therapy, however, failed to demonstrate superior results to thrombolytic treatment alone. After thrombolytic treatment only 60% of patients have a full restoration of coronary blood flow (TIMI 3), whereas 20–30% still have reduced flow of TIMI 2. Using PTCA for reopening of occluded coronary vessels, TIMI flow 3 is achieved in 90% of patients. In the beginning of the thrombolytic area, vessels with TIMI flow 2 and 3 have been regarded as patent vessels and this criterion has been used to describe the efficacy of thrombolytic agents. However, patients with reduced coronary blood flow (TIMI 2) have reduced prognosis compared to patients with TIMI flow 3.19 This prognosis is as poor as if the vessels had never been opened. Using contrast echocardiography, it has been demonstrated that, despite reopening of coronary vessels in acute myocardial infarction, full restoration of myocardial perfusion is incomplete in parts of the patient’s heart. Areas with incomplete restoration of flow and reduced or no contrast enhancement of the myocardium in echocardiography have less improvement of ventricular function than areas with complete opacification.20 Full contrast opacification after reopening of the infarct related vessel is an indicator of viable myocardium with functional improvement during follow up. Intracoronary Doppler flow velocity measurements demonstrate an increase in coronary flow velocity reserve in recanalised infarcted arteries which is also related to left ventricular recovery.
The incomplete restoration of ventricular function and perfusion may be explained by the so called “no reflow” phenomenon resulting from PTCA—induced microthromboemboli, which may reverse spontaneously.21 Additionally, differences in collateral blood flow may play a role. Also, abnormalities of microvascular perfusion secondary to leucocyte plugging and microembolisation must be discussed.22 An increased expression of neutrophil and monocyte adhesion molecules could be shown. They can induce enhanced vasoconstriction, microvasculature functional abnormalities caused by serotonin, thromboxane A2 or other substances, and induce local thombotic effects caused by tissue factor expression secondary to monocyte adhesion. Activated leucocytes may form microaggregates, causing plugging in the microvasculature. New therapeutic options may be based on these findings.
After reopening of coronary vessels in acute myocardial infarction, recurrent chest pain as well as ST segment elevation can be observed in 10–15% of patients. Patients with this phenomenon have elevated creatine kinase, lower ejection fraction, and a worse prognosis than patients with a rapid decline of the ST segment after reopening of the vessel. This is also true for those patients who have permanent ST segment elevation compared to those with a transient ST shift. Permanent ST segment elevation is associated with extensive infarction and reduced recovery of ventricular function.
Coronary blood flow and ventricular function can be improved with glycoprotein IIb/IIIa antagonist treatment to reduce thrombus formation and fragmentation at the lesion site with subsequent distal microembolisation, resulting in improved myocardial perfusion.23
These new clinical findings indicate the importance of the coronary microcirculation for clinical cardiology. New methodological techniques to study the coronary microcirculation have demonstrated an interaction between epicardial arteries and the microcirculation. Functional and morphological changes are observed in patients treated for coronary artery disease by interventional cardiology. Nearly 50% of the patients have reduced coronary flow reserve in the reference vessel—that is, microvasculature disease which is present even when no significant coronary luminal narrowing is evident. An interesting explanation may be the presence of spontaneous or interventionally induced microembolisation after plaque rupture, which can be repetitive, recurrent, and lead to a significant reduction of myocardial and microvasculature function even with normal or nearly normal coronary arteries.24–26 Also, for ischaemic cardiomyopathy the pathogenesis could be repetitive coronary microembolisation after multiple plaque rupture and healing, described as multilayered intimal thickening. Inflammation detected by elevation of C reactive proteins or cytokines can in part be the result of ongoing microembolisation.