Cerebral hyperperfusion, or reperfusion syndrome, is a rare, but serious, complication following revascularization. Hyperperfusion is defined as a major increase in ipsilateral cerebral blood flow (CBF) that is well above the metabolic demands of the brain tissue. Quantitatively, hyperperfusion is a 100% or greater increase in CBF compared with baseline.[1]
This definition also extends to rapid restoration of normal perfusion pressure, for example, with thrombolytic therapy for acute ischemic stroke. Reperfusion syndrome can occur as a complication of carotid endarterectomy (CEA), intracranial stenting, and even bland cerebral infarction.
The terms hyperperfusion and reperfusion are often used interchangeably. The former implies excessive flow, while the later suggests normalization of flow.[1, 2] Both can result in cerebral injury with similar clinical pictures, which is the reason for the substitution of terms. However, not all patients with hyperperfusion are symptomatic; conversely, patients with only moderate rises in CBF can have devastating outcomes. Therefore, some authors prefer to address this subject as reperfusion syndrome.[2]
When patients are identified and treated early, the prognosis is better and the incidence of intracranial hemorrhage is decreased.[3] Outcomes are dependent on timely recognition and prevention of precipitating factors. Most important is the treatment of hypertension before it can inflict damage in the form of edema or hemorrhage.
The prognosis following hemorrhagic transformation is poor. Mortality in such cases is 36-63%, and 80% of survivors have significant morbidity.[4, 5, 6]
Studies indicate that reperfusion injury is involved directly in the potentiation of stroke damage. Components of the inflammatory response, including cytokine release and leukocyte adhesion, appear to play key roles in these deleterious effects.
Damage to the blood-brain barrier (BBB), an important factor in reperfusion injury, is seen in the image below.
View Image | Postcontrast image 24 hours after a right middle cerebral artery stroke, demonstrating contrast extravasation through a faulty blood-brain barrier. |
Cerebral reperfusion syndrome presents as a triad of ipsilateral headache, contralateral neurological deficits, and seizure.[1]
The time frame in which symptoms arise can be from immediately after restoration of blood flow to up to 1 month after restoration. Patients are usually symptomatic within the first week.[1, 4, 7]
Headache is the most common symptom (62%).[8] Typically, patients display migrainous features with severe, ipsilateral, pounding headache.
Deficits are usually cortical (eg, hemiplegia, neglect, aphasia) or may involve worsening of a preexisting deficit. By the same token, seizures may present as focal or generalized, depending on the cortical area affected.[1, 7]
Several mechanisms have been proposed for the pathogenesis of cerebral reperfusion injury. As time passes following arterial occlusion or partial occlusion, the basic idea is that for a period of time collateral circulation will sustain normal neurological function, which is dependent on the individual and their risk factors. Eventually compensation for hypoperfusion will lead to increased vascular resistance and venous collapse.[9] The extent of reperfusion injury will depend on the individuals time since collateral collapse and irreversibly damaged tissues. There are a number of events that can lead to this situation, such as postoperative hypertension, to molecular modalities, such as free oxygen radical release. Each theory is complex and none are widely accepted. For the time being, known risk factors include the following[10] :
Elevated blood pressure is the most common factor found in symptomatic patients.[8, 10, 11] During acute ischemic stroke, systemic blood pressure often rises as a physiologic compensation for cerebral ischemia.[12] As a rule, elevated blood pressure is not treated, so as not to compromise flow to the tenuous penumbra. The key to reperfusion injury in this scenario is ischemic disruption of the blood-brain barrier (BBB). The offended BBB contains abnormally permeable ischemic capillaries.
Adding insult to injury, these small vessels do not have a substantial conduit to buffer systemic pressures. The injured endothelium is unable to maintain its structural integrity against systemic vascular resistance, thus resulting in reperfusion injury or hemorrhagic transformation. Hypertension-related hemorrhage is seen in the image below.[13]
View Image | T1 sagittal image without contrast demonstrating gyriform hyperintensities. These represent subacute petechial hemorrhage around an area of subacute i.... |
Cerebral autoregulation protects the brain against changes in systemic blood pressure. A drop in blood pressure could lead to ischemia, while on the other hand, a sudden rise could lead to edema or hemorrhage. In patients with high-grade stenosis, CBF is maintained at the expense of maximal arteriolar vasodilatation.[14]
Chronic cerebral hypoperfusion (eg, critical stenosis) leads to the production of carbon dioxide and nitric oxide. These are vasodilatory substances that cause endothelial dysfunction.[15] In the absence of cerebral autoregulation, CBF is directly dependent on the systemic blood pressure. Correction of a critical stenosis causes rapid and large changes in the CBF, which can lead to edema or hemorrhage.[16]
Ischemia-reperfusion injury is characterized by oxidant production, complement activation, and increased microvascular permeability. Various cytokines peak in the serum within the first 24 hours of an acute stroke and are thought to initiate the cascade of tissue damage. At the site of ischemia itself, activated leukocytes release free radicals and toxins, causing further destruction. The combination results in an impaired BBB, which can lead to cerebral edema and/or hemorrhage.[17] These changes are especially important in the setting of hypertension, as indicated above.
Symptomatic hemorrhagic transformation rates within 24-36 hours of stroke are increased in the setting of revascularization therapy, regardless of modality (ie, intravenous lytics, intra-arterial lytics, antithrombotics, or mechanical devices).[18] In the absence of revascularization therapy, hemorrhagic transformation is a common and natural consequence of infarction.[19, 20]
In the setting of revascularization, the fundamental question is whether the increased rates of hemorrhagic transformation are caused by reperfusion and the biochemical pathways, or if they are specific consequences of the lytic state itself.
Rates of symptomatic intracerebral hemorrhage are generally higher in intra-arterial lytic trials[21] (eg, 10% in PROACT-II) than in intravenous lytic trials (eg, 6.4% in NINDS).[22] The rates of symptomatic ICH following revascularization with a device are even lower and range from 4%-2% with the Trevo and Solitaire stent systems, respectively.[23, 24]
Hemorrhagic transformation is now known to be a multifactorial process. Stroke severity is likely to be a major predictor of symptomatic intracerebral hemorrhage because it is associated with the volume of ischemic brain at risk for hemorrhagic transformation. Older patients may be at greater risk of symptomatic intracerebral hemorrhage.
Higher lytic doses are associated with higher symptomatic intracerebral hemorrhage risk, but whether lower doses can achieve adequate benefit with less risk is not known. Delayed revascularization minimizes benefit and likely increases risk. The goal of acute revascularization should not just be to open occluded vessels but to open them quickly.
Patient selection based on physiologic parameters is likely important to reduce late hemorrhage attributable to revascularization.
The use of readily available perfusion imaging to screen for potentially salvageable brain tissue, faster door-to-recanalization times, and the introduction of next generation stent retriever devices led to the completion and publication of six positive endovascular mechanical thrombectomy trials in 2015 and 2016, making it the new standard of care for acute ischemic strokes resulting from large artery occlusion in the anterior circulation.[25, 26, 27, 28, 29, 30]
Evolving acute ischemic strokes resulting from proximal cerebral artery occlusions often result in large areas of penumbra that are likely vulnerable to reperfusion injury following revascularization. Longer periods of hypoperfusion to these areas, incurred as the therapeutic treatment window has extended beyond 6 hours in patients with favorable perfusion imaging (see image below), could possibly result in increased susceptibility of recoverable tissue to additional injury.
It currently remains unclear to what extent reperfusion injury might contribute to loss of penumbra following mechanical thrombectomy. Further research in this area could have the potential to improve post-thrombectomy outcomes through its prevention and treatment.
Two studies have demonstrated that higher blood pressure goals following thrombectomy, traditionally thought to be beneficial in acute stroke by promoting perfusion of the penumbra, correlated with worse clinical outcomes.[31, 32] It is possible that these findings are in part due to the negative effects of reperfusion injury. Neuroprotective agents, whose previous failures in clinical trials could have resulted from insufficient delivery to tissues at risk for irreversible injury behind occluded vessels, might also deserve a second look.
View Image | CT perfusion scan. Matched image sets with irreversible injury shown in magenta on the left and at-risk tissue (penumbra) shown in green on the right. |
Transcranial Doppler (TCD) ultrasonography measures cerebral blood flow in major cerebral arteries. Low preoperative distal carotid artery pressure (< 40 mm Hg) and an increased peak blood flow velocity have been found to be predictive of postoperative hyperperfusion.[16, 17] Therefore, TCD can be used to select patients for aggressive postprocedure observation and management. In a patient who is determined to be at risk, TCD can also be used during the postoperative period to assess for hyperperfusion.
Cerebrovascular reactivity (CVR) to carbon dioxide can be used to test cerebral hemodynamic reserve. Normally, administration of acetazolamide (a carbonic anhydrase inhibitor that causes a local increase in carbon dioxide) induces a rapid increase in CBF.[33] This iatrogenic CBF surge is measured using single-photon emission computed tomography (SPECT) scanning.
In chronic cerebral ischemia, the vasculature is maximally dilated. Therefore, there is little change in CBF, which means decreased CVR. Patients with low preoperative CVR are at risk for developing hyperperfusion and, thus, parenchymal injury.[3, 34]
The most important factor in preventing reperfusion syndrome is early identification and control of hypertension.[4, 5, 6] This is important even in normotensive patients, since delayed hypertension can occur.[7] The use of TCD ultrasonography preoperatively and postoperatively can aid in identifying patients with increased CBF and, consequently, increased risk of hyperperfusion.[3, 34] Blood pressure should then be controlled aggressively if CBF elevates.
In the situation of reperfusion after carotid endarterectomy (CEA), Cleveland Clinic has implemented an effective protocol for identifying risk factors of reperfusion syndrome and post-op hemorrhage. These risk factors include stenosis of >80%, pre-morbid hypertension, and poor collaterals. In these patients the BP was maintained at less than 120/80 post-op. Of 225 patients, 33% (n=75), 0 patients had post-op reperfusion syndrome, or hemorrhage. Prior to the protocol implementation they had a 17% complication rate.
Pressures can be reduced gently with antihypertensives that do not increase CBF or cause excessive vasodilatation. Examples include labetalol (Normodyne, Trandate) and nicardipine (Cardene). Less favored medications include intravenous angiotensin-converting enzyme (ACE) inhibitors, calcium-channel blockers, and vasodilators such as nitroprusside.[3, 35]
Unfortunately, no specific parameters or guidelines have yet been established for optimal blood pressure under these circumstances. According to the American Stroke Association stroke and intracerebral hemorrhage guidelines, the blood pressure goal for an acute intracerebral hemorrhage is a mean arterial pressure (MAP) of less than 110 mm Hg.[36, 37] This modest pressure goal can also be applied in acute ischemic stroke with reperfusion issues, because it does not hypoperfuse the tenuous surrounding tissues, nor does it further aggravate injury or hemorrhagic conversion.
In any case, it remains the consensus that patients should be observed postoperatively in an intensive care unit (ICU) setting. If blood pressure management is an issue, it should be managed in the ICU until stabilized.
Free radicals produced during ischemia are a purported culprit in reperfusion injury. Free-radical scavengers and antiadhesion therapy have shown promise in decreasing the incidence of endothelial injury.[15]
Animal studies using various methods of modulating the cytokine response have shown beneficial effects from modulation of IL-1 and TNF. Various experimental studies using agents that block leukocyte endothelial adhesion (ie, monoclonal antibodies that block either the adhesion receptor on leukocytes [CD-18] or the corresponding adhesion receptor on the endothelial cell [ICAM-1]) have shown beneficial effects in terms of reducing infarct size and improving functional outcome.[38] The adhesion process and the mechanism behind antiadhesion therapy are illustrated in the image below.
View Image | A. Schematic representation of the process of endothelial-dependent leukocyte adhesion. Endothelial cells activated by histamine or thrombin rapidly t.... |
Other molecules that have been tested and found to be successful in neuroprotection in mice or rats but not humans include, a-lipoic acid,[39] magnolol,[40] and Punicalagin.[41] Another finding that might explain the difficulty of translating from murine to human is that during times of sublethal stress thousands of genes may be silenced in a process called preconditioning. The preconditioning sets up multiple neuroprotective pathways and results in attenuation of the subsequent ischemic insult.[42] Therefore, rather than a single molecule for neuroprotection, we will likely need a multimodal cocktail of molecules to save patients from pending ischemia and neuronal death.
Natalizumab, a monoclonal antibody against an adhesion molecule on white blood cells that inhibits their movement across the blood-brain barrier, already proven to be an effective treatment in multiple sclerosis, completed a phase 2 trial in patients with acute ischemic stroke. Though no effect on reducing the increase in irreversible tissue injury typically following the acute phase of an ischemic stroke was demonstrated, there was some possible benefit to function outcomes in patients who received treatment.[43] A larger study using natalizumab in acute stroke is currently in progress.
In general, these experimental studies have shown benefit when a period of ischemia is followed by a return of blood flow (reperfusion), but not when ischemia is permanent. For this reason, antiadhesion therapies may prove to be most beneficial clinically when given in association with thrombolytic agents or thrombectomy.
Although clinical studies using antibodies against ICAM-1 have failed to show a clinical benefit, further investigations of antiadhesion therapies in combination with t-PA and thrombectomy are ongoing. Given the strong preclinical evidence for the usefulness of anti–reperfusion injury agents, such agents are likely to be used in future "stroke cocktail" therapeutic efforts.
Researchers assessed the use of stem cell product MultiStem in adults with ischemic stroke. Patients who received MultiStem within 36 hours of symptom onset demonstrated better global outcomes than those who received placebo.[44] It is thought that some of the benefits seen in this trial might be related to a decrease in the cell-mediated production of inflammatory mediators that typically follows acute ischemic strokes.[45] A phase III trial using the MultiStem cell product in acute stroke is set to begin enrolling the summer of 2018.
A. Schematic representation of the process of endothelial-dependent leukocyte adhesion. Endothelial cells activated by histamine or thrombin rapidly translocate P-selectin to their surfaces (also E-selectin, not shown), tethering leukocytes to the endothelial cell. This tethering does not require an active response from the leukocyte. Once tethered, other factors, including platelet-activating factor and cytokines, are released to stimulate a leukocyte activation response. This response includes shape-changing and increased surface expression of CD-11/CD-18. CD-11/CD-18 then binds to the corresponding intercellular adhesion molecule 1 (ICAM-1) receptor on the endothelial cell, leading to firm endothelial attachment. This attachment may produce direct obstruction of the microcirculation or lead to infiltration into the surrounding brain parenchyma. B. Schematic representation showing that through the use of monoclonal antibodies directed against the ICAM-1 receptor, the CD-11/CD-18 to ICAM-1 attachment is prevented. This, in turn, prevents subsequent microvascular obstruction and leukocyte infiltration.
A. Schematic representation of the process of endothelial-dependent leukocyte adhesion. Endothelial cells activated by histamine or thrombin rapidly translocate P-selectin to their surfaces (also E-selectin, not shown), tethering leukocytes to the endothelial cell. This tethering does not require an active response from the leukocyte. Once tethered, other factors, including platelet-activating factor and cytokines, are released to stimulate a leukocyte activation response. This response includes shape-changing and increased surface expression of CD-11/CD-18. CD-11/CD-18 then binds to the corresponding intercellular adhesion molecule 1 (ICAM-1) receptor on the endothelial cell, leading to firm endothelial attachment. This attachment may produce direct obstruction of the microcirculation or lead to infiltration into the surrounding brain parenchyma. B. Schematic representation showing that through the use of monoclonal antibodies directed against the ICAM-1 receptor, the CD-11/CD-18 to ICAM-1 attachment is prevented. This, in turn, prevents subsequent microvascular obstruction and leukocyte infiltration.