RETROGRADE TRANSVENOUS PERFUSION - 08/09/11
Riassunto |
Early effective treatment for stroke has been a quest of investigators and clinicians for decades. Traditional surgical therapies have been used to prevent strokes. These would include carotid endarterectomy and extracranial to intracranial arterial bypass. With the exception of acute embolectomy, surgical treatment has not been effective for acute stroke and could in some instances produce a more severe injury from reperfusion of the ischemic bed. Surgical therapies are generally slow to perform and are therefore unsuitable for emergency stroke treatment. This has given rise to new therapies such as tissue plasminogen activator (tPA) delivered intravenously or intra-arterially. These therapies are quickly started and could potentially benefit the majority of ischemic stroke patients because the cerebral vessel occlusion is most commonly caused by a blood clot. However, they do have important drawbacks. Although the time necessary to start intravenous thrombolysis with tPA is very short, the number of patients who arrive in the emergency room within the 3-hour therapeutic window is small and treatment after this window results in a higher risk of intracerebral bleeding. Intra-arterial thrombolysis requires time to catheterize the appropriate cerebral vessel and commonly takes 1 to 2 hours to lyse the clot. Both intravenous and intra-arterial therapies carry a risk of intraparenchymal bleeding. Therefore, there is room for another approach to acute stroke treatment.
In our search for a highly effective stroke treatment, we began to look outside the common approaches of cytoprotective agents and clot lysis drugs. We sought an approach that would quickly supply oxygen-rich blood to ischemic tissues. We also sought a method that would carry a low risk of hemorrhage and brain edema and that would permit the delivery of cytoprotective drugs. The desired technique would be capable of prolonged ischemic tissue support. We were particularly interested in a therapy that would provide rapid, measurable improvement (within hours) in a substantial portion (50% or more) of treated patients.
In the mid-1980s, reports appeared describing a new technique in which autologous arterial blood is pumped in a retrograde direction through the coronary sinus for the reversal of myocardial ischemia. The intent of this therapy was to provide arterial blood directly to the ischemic myocardium until the arterial obstruction could be cleared or bypassed. Reports from human trials suggested that it was both safe and effective.7
Early literature contained reports of surgical procedures designed to direct arterial blood retrograde to brain tissue. The earliest of these attempts were laboratory experiments in which the carotid artery and jugular vein were anastomosed.3 Unfortunately, the arterial blood was not directed retrograde toward the brain but instead passed directly back to the heart. Refining the technique by occluding the ipsilateral jugular vein, Beck et al1 reported improvement in children who had mental and convulsive disorders. However, the following year, Gurdjian et al8 reported that there was no significant improvement in oxygenation of the superior sagittal sinus when this technique was used in monkeys. No further work for cerebral retrograde perfusion seems to have been done until reports began to appear in the cardiothoracic literature. Surgeons reported small experiences with completely reversing the flow of arterial blood during open-heart surgery to wash out air and particulate emboli.9 These historical reports of retrograde perfusion of the brain and the coronary sinus experiences led us to adapt and refine the coronary retroperfusion technique into what is now called retrograde transvenous neuroperfusion (RTN).
Beginning in 1986 at UCLA, it took 11 years to develop this technique and involved more than 100 primate experiments. In its simplest form, RTN was meant to provide arterial blood rapidly and temporarily to ischemic brain tissue until the original arterial obstruction spontaneously lysed, could be cleared with lytic therapy, or could be bypassed. The technique involved harvesting autologous blood from a femoral artery with an external pump and passing that blood through catheters located in the transverse sinuses of the brain. From there, the blood would pass into the veins of the brain and capillary bed (Figure 1). It became obvious during experiments that catheter tip placement was crucial to the success of the procedure. Catheters located too far lateral obstructed crucial, alternate venous outflow through the veins of Labbe and the superior petrosal sinuses. When the balloons on the tips of the laterally located RTN catheters were inflated for partial obstruction of the outflow of blood from the transverse sinuses, they also obstructed these collateral pathways, resulting in high-resting venous pressures and even higher pressures when RTN flow was started. It was discovered that simply moving the catheters medially near the torcular resulted in a dramatic drop in resting venous sinus pressures to near normal. When RTN was then started, pressures of only 10 to 20 mm Hg were necessary to move blood retrograde into the venous system.
It was recognized that forcing blood to flow retrograde through veins and venules to the capillary bed was not an intuitively appealing concept. The first concern voiced by critics of RTN was for valves in the venous system that would obstruct the retrograde flow of blood. Thankfully, there are no valves in the cerebral venous system. The only valves that might cause obstruction are present in the proximal jugular system. Therefore, retrograde perfusion from the superior vena cava, as has been done in some cardiac surgery cases, may not reach the brain. This argues for the placement of catheters intracranially.
A second and much greater concern of critics was that the superior sagittal sinus is the major pathway for flow of blood from the brain. Obstructing this path, it was argued, would lead to serious consequences. This, in part, stemmed from the historical observation that superior sagittal sinus occlusion was accompanied by cerebral hemorrhage. This was underscored by neurosurgeons who found that occlusion of the superior sagittal sinus during surgery was accompanied by venous congestion and hemorrhagic infarction. Therefore, it was reasoned that forcing blood to flow through the superior sagittal sinus in a retrograde manner would likely cause similar consequences. Hemorrhagic infarction was attributed to two factors. One, that a pressure greater than normal sinus pressure would be needed to bring about the retrograde flow. In turn, there would be pressure elevation in the venous and capillary beds, leading to hemorrhage from such fragile vessels. Two, an obstruction of the major venous outflow from the brain would lead to severe congestion, elevated capillary bed pressures from the continued arterial inflow, and ultimately cerebral hemorrhage. What has more recently become apparent is how biased our sampling of patients who have superior sinus occlusion has been. We, in fact, only see those patients who are symptomatic from their sinus occlusion. It appears that a much larger proportion of patients who have sinus occlusion remain asymptomatic.2 This suggests that the superior sagittal sinus can in fact be occluded without serious consequences. It could be argued that the sinus occlusion may form slowly, allowing time for collateral venous outflow to develop, and, therefore, patients remain asymptomatic. However, we believe that the collateral venous network is large and quite redundant and can readily handle the change in the outflow pattern that occurs when RTN forces blood through the superior sagittal sinus in a retrograde fashion. This concept is supported by the venous phase of angiography. Angiography shows such collateral paths that include the cavernous sinus, the basilar plexus, and the veins of Labbe.
It may also be that RTN and acute sinus occlusion are actually quite different and therefore, not comparable. Acute sinus occlusion promotes stasis and clotting of blood within the sinus and connected veins and could indeed lead to hemorrhagic infarction in those areas with poor collateral venous outflow. On the other hand, RTN can be seen as promoting continued flow, only in a retrograde direction. RTN does not promote stasis in areas with poor venous collateral flow. Instead, it helps to redirect flow from those areas to the collateral sites of outflow mentioned above.
There is additional experimental data to suggest that RTN does not require unusual pressures to force blood in a retrograde manner into the ischemic capillary bed. Symon10 measured middle cerebral artery (MCA) pressures in primates before and after MCA occlusion. Before occlusion, the pressure measured in the proximal MCA averaged 94 mm Hg, and middle cerebral venous (MCV) pressures averaged 14 mm Hg. If one assumes that there is an approximately 80% drop in pressure across the arteriolar bed, then the normal capillary bed pressure should average 18 to 28 mm Hg. This correlates well with the 18 to 22 mm Hg actual measurements of an accessible capillary bed in the finger.4 It would therefore be predicted that when RTN reached pressures near 20 mm Hg, blood would flow retrograde into a normal capillary bed.
When Symon10 measured pressures during occlusion of the MCA, he found that the mean arterial pressure distal to the occulsion was 20.6 mm Hg and that MCV pressures were 9.4 mm Hg. The pressure in the capillary bed would be between these two and closest to the venous pressure caused by the drop in pressure across the arteriolar bed. Thus, RTN pressures would likely need to be only slightly higher than the venous pressures (9 mm Hg) for flow to move into the ischemic capillary bed. These pressures are in keeping with our own experimental results in baboons.5
There are additional primate experimental results that help to demonstrate that RTN can move arterial blood to the ischemic brain without hemorrhage or venous congestion. Early RTN research concentrated on venous angiography. It became apparent that dramatic filling of venous sinuses, along with deep and superficial veins, was possible when the pressure of retrograde perfusion was 30 mm Hg or greater (Figure 2). However, these pressures were thought to be higher than was actually necessary to transport blood retrograde to the capillary bed as indicated above. Studies were performed that allowed for the direct observation of the cortical surface in both the ischemic and the nonischemic hemispheres. It was possible to see that retrograde flow occurred in the surface veins at much lower pressures and that cortical capillary blood flow, as measured by laser Doppler, was increased to greater than normal levels in the ischemic area. Therefore, it was presumed that venous angiography did not fully demonstrate the penetration of RTN.
Additional early primate experiments demonstrated the penetration into brain tissue of RTN-transported dye. During occlusion of the right MCA and with RTN being performed, red dye was injected into the arterial system and blue dye was mixed with the retroperfused blood. After approximately 30 seconds, the animal was sacrificed. These studies demonstrated excellent penetration of blue, RTN-perfused dye into the ischemic middle cerebral territory (Figure 3). Microscopic studies confirmed penetration to the capillary bed (Figure 4). There was no evidence of dye leaking from the capillaries, hemorrhage, or brain edema.
The final evidence for both the safety and efficacy of RTN came from a series of 10 baboon survival experiments. The primates were divided into two equal groups and subjected to MCA occlusion for 3.5 hours after which time the animals were recovered, followed for 6 days, and then sacrificed. All animals were treated identically with the exception that five received RTN during the last 2.5 hours of occlusion. The results of somatosensory-evoked potentials (SSEPs) during the ischemic period, clinical outcome during the 6 days of observation, and size of brain infarction were compared for the two groups.5 The treated group clearly fared better than the untreated group. SSEPs, which demonstrated ischemia during the first hour of MCA occlusion, recovered to better than normal with RTN (103%) compared with 75% in the untreated group (P < 0.01%). Clinical outcome scores were graded on a scale of 0 to 100, with the latter considered normal. Treated animals had an average score of 99, whereas the score for the untreated group was 66 (P < 0.015). The average infarction volume was 0.02% of the ipsilateral hemisphere in the RTN-treated group and 3.9% in the untreated group (P < 0.05). These data led us to propose opening a clinical trial at UCLA.
Il testo completo di questo articolo è disponibile in PDF.Mappa
| Address reprint requests to John G. Frazee, MD, Division of Neurosurgery, Rm 18-211, NPI, UCLA School of Medicine, Box 957039, Los Angeles, CA 90095-7039 |
Vol 15 - N° 4
P. 777-788 - ottobre 1999 Ritorno al numero
Articolo precedente
- NEUROENDOVASCULAR RESCUE : Nonthrombolytic Approach To Acute Brain Ischemia
- Camilo R. Gomez, R. Van Wadlington, John B. Terry, Roekchai Tulyapronchote
- ENDOVASCULAR THERAPY OF TRAUMATIC INJURIES OF THE EXTRACRANIAL CEREBRAL ARTERIES
- Camilo R. Gomez, Addison K. May, John B. Terry, Roekchai Tulyapronchote
Benvenuto su EM|consulte, il riferimento dei professionisti della salute.
L'accesso al testo integrale di questo articolo richiede un abbonamento.
Già abbonato a @@106933@@ rivista ?