No-Reflow Phenomenon. A New Target for Therapy of Acute Myocardial Infarction Independent of Myocardial Infarct Size

The no-reflow phenomenon is a failure to reperfuse portions of previously ischemic tissue when patency of the large arteries perfusing an organ has been reestablished. This phenomenon has been described in many organs but is very common in both animal models and humans after restoration of epicardial coronary artery patency in the setting of ST elevation myocardial infarction. ¹ In early experimental studies carried out in the anesthetized canine model, a 90-minute proximal circumflex coronary artery occlusion (induced mechanically by a clamp on the coronary artery) followed by reestablishing patency (releasing the clamp) demonstrated persistent perfusion defects in the subendocardial myocardium, assessed by injection of a fluorescent dye into the vasculature. Histologic and ultrastructural analysis of the tissue within these no-reflow zones demonstrated microvascular obstruction. ² The predominant ultrastructural finding was localized endothelial swelling resembling blebs or blisters that appeared to be clogging the lumen of small blood vessels including the capillaries. The endothelium also demonstrated a reduction in pinocytotic vesicles and the presence of endothelial rupture. Platelet deposition, fibrin tactoids, microscopic hemorrhage, leukocyte plugging, and rouleaux formation were less commonly observed. Occasionally, capillaries appeared to be compressed by swollen myocytes. ² The no-reflow zones, assessed by lack of the fluorescent dye and other markers such as carbon black, appeared to be confined well within areas of already irreversibly injured myocytes. Ultrastructural evidence of myocardial cell death appeared before ultrastructural evidence of microvascular damage. ³ It is our opinion that no-reflow does not directly contribute to myocyte cell death. However, because flow into and out of the necrotic zone is inhibited by no-reflow, removal of necrotic debris from the infarct zone may be hampered by no-reflow, and cytokines and cells involved in the healing phase of the myocardial infarct likely have difficulty getting into and out of the area. Both experimental and clinical studies have shown that no-reflow is associated with poor healing of the infarct, with resulting worsened postmyocardial infarction adverse left ventricular (LV) remodeling, including thinning and stretching of the infarct (infarct expansion), dilatation of the LV cavity, and worsened LV function, as well as increased mortality. ¹ Of note, the negative effects of no-reflow on clinical outcomes have been shown in several studies to be independent of the size of the infarction. ^4,5 If that is so, then therapy aimed specifically at reducing no-reflow, and not necessarily infarct size, makes sense. ⁶

How does one reduce no-reflow? In many cases, reducing myocardial infarct size will in fact reduce no-reflow. The smaller the infarct, the smaller the zone of no-reflow in many but not all situations. An intervention can have variable effects on necrosis and no-reflow (Table 1). We have observed that the no-reflow zone is smaller when the infarct is reduced by earlier reperfusion, ischemic preconditioning, ¹⁰ cariporide, a sodium potassium exchange inhibitor, ¹⁰ and hypothermia induced during the ischemic episode. ⁸ However, there are some pharmacologic agents that reduce infarct size in our experimental models but fail to reduce the zone of no-reflow; for example, the mitochondrial protective agent, SBT-20 ¹² and the antianginal agent, ranolazine. ¹⁴ In addition, we have found some therapies that reduce no-reflow but have no effect on infarct size. An example that we have explored in 2 animal models (rabbit and rat) is therapeutic hypothermia started after reperfusion. ^15,16 Cooling the heart after reperfusion (1 minute after in the rat model; 5 or 30 minutes after in the rabbit model) significantly reduced no-reflow but had no effect on infarct size. We are currently carrying out a study to determine whether hypothermia induced after reperfusion, which is known to reduce no-reflow but not infarct size, will also reduce adverse LV remodeling. Our group previously showed that treatment with oxygen radical scavengers at reperfusion decreased microvascular damage and improved regional coronary artery blood flow, without reducing myocardial infarct size. ^19,24 A recent preliminary study showed that ischemic preconditioning did not reduce infarct size in a mouse model of prolonged coronary occlusion and reperfusion but did reduce no-reflow, and this was associated with less adverse LV remodeling. ²⁵ Of note, neither thrombolytic therapies nor anticoagulant therapies have reduced no-reflow size in our experimental models.

Table 1.

Effects of Various Interventions on Myocardial Infarct Size (IS) and No-Reflow (NR).^a

Intervention	Reference	Effect on IS	Effect on NR
Agents that reduce both IS and NR
Hypothermia (started before reperfusion, rabbit model)	7	↓	↓
Hypothermia (started before reperfusion, rat model)	8	↓	↓
GP531, (adenosine-regulating agent, rabbit model)	9	↓	↓
Cariporide (given before reperfusion, rabbit model)	10	↓	↓
Ischemic preconditioning (rabbit model)	10	↓	↓
Early reperfusion (canine model)	2	↓	↓
Agent that increases both
Hyperthermia (rabbit model)	11	↑	↑
Agents that reduce IS but not NR
SBT-20 (given before reperfusion, rat model)	12	↓	–
Olmesartan (given before reperfusion, rat model)	13	↓	–
Ranolazine (rabbit model)	14	↓	–
Agents that reduce NR but not IS
Hypothermia (started after reperfusion, rabbit model)	15	–	↓
Hypothermia (started after reperfusion, rat model)	16	–	↓
Diannexin (large ischemic risk zones, rabbit model)	17	–	↓
Bendavia (given before reperfusion, rabbit model)	18	–	↓
Superoxide dismutase + catalase (canine model)	19	–	↓
Agents that do not affect IS or NR
Dabigatran (rabbit model)	20	–	–
Heliox breathing (rabbit model)	21	–	–
Tissue-type plasminogen activator (canine model)	22	–	–
Streptokinase (canine model)	23	–	–

Abbreviations: IS, infarct size; NR, no-reflow.

^a→No effect, ↓ reduced damage, ↑ increased damage. Table 1 represents studies performed in the Kloner laboratory over several decades. Involves mechanical occlusion/reperfusion of the epicardial infarct-related coronary artery.

Adverse LV remodeling following ST elevation myocardial infarction is a major determinant of prognosis. In general, the more dilated the left ventricle becomes in the chronic stages of myocardial infarction, the worse the clinical outcome and the higher the mortality. ²⁶ The process of LV remodeling begins within the first week of myocardial infarction but then extends for months. Initially, when the interstitium breaks down, necrotic myocytes slide past each other and the necrotic zone stretches and thins. This process is known as infarct expansion. ^27,28 It is associated with regional LV dilatation followed by global LV dilatation. Apoptosis occurring at the border of the infarct may also contribute. The LV may continue to dilate over the course of many months. Eccentric hypertrophy of the noninfarcted LV wall in which sarcomeres are added in a longitudinal fashion also contributes to the LV dilatation. The treatment for LV remodeling has largely been with pharmacologic therapies such as angiotensin converting enzyme inhibitor or angiotensin receptor blocker or beta-blockers. These agents can be started as late as a week after infarction and, when given chronically after myocardial infarction, reduce LV remodeling (likely in part by afterload and preload reduction) and improve survival. ²⁹ However, reduction in major adverse events is usually on the order of ∼20%. That suggests that there is still a large unmet need to further reduce adverse LV remodeling. Whereas reperfusion reduces infarct size and this alone can reduce LV remodeling, remodeling still occurs, even when infarcts are successfully reperfused. Here is where treatment of no-reflow could further improve outcome—by improving the ability of the infarct to heal. The treatment could be started after patency of the epicardial coronary artery has been established, using a therapy such as cardiac hypothermia, and we predict that by reducing no-reflow this will reduce adverse LV remodeling. By starting therapy after reperfusion, there would be no delay in door to balloon/stent time, which is an issue when adjunctive therapy has to be started prior to reperfusion therapy.

Current therapy for myocardial infarction relies heavily on opening the epicardial coronary artery as soon as possible and keeping it open. Mortality rates for ST elevation myocardial infarction have fallen as door to balloon/stent times have fallen, but a recent analysis showed that these mortality rates are plateauing. ³⁰ Early reperfusion works primarily because it reduces myocardial infarct size. But attempts to further reduce myocardial infarct size with adjunctive pharmacologic therapy along with reperfusion have been mixed ³¹ and there is no accepted standard adjunctive therapy to further reduce infarct size above and beyond early reperfusion. However, treating the no-reflow phenomenon independently of reducing myocardial infarct size offers another potential therapy for improving the outcome of ST elevation myocardial infarction. Treating no-reflow should improve healing of the infarct, reduce adverse LV remodeling, improve LV function, and perhaps reduce heart failure and mortality. There is a need for future research on therapies that reduce the no-reflow phenomenon independent of the effect on myocardial infarct size.