Microsurgical Embolectomy in the Current Era of Pharmacological and Mechanical (Endovascular) Thrombolysis—A Reappraisal
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.319226
Source of Support: None, Conflict of Interest: None
Keywords: Acute ischemic stroke, acute large vessel occlusion, arteriotomy, endovascular embolectomy contraindicated/failure/complication, surgical embolectomyKey Message: Microsurgical embolectomy remains an indispensable recanalization tool and should not be dismissed as an antiquated reperfusion method.
The journey of vascular recanalization for “acute embolic stroke due to large vessel occlusion” started somewhere in mid-1950s with surgical embolectomy as its oldest treatment strategy. The first report of two cases of open embolectomy for Middle Cerebral Artery Occlusion (MCAO) was presented by Welch at a meeting of Harvey Cushing Society in Mexico in 1955 and then published in 1956. Though surgical embolectomy gradually gained acceptance with publication of more reports,,,, it never becomes an ideal mainstream procedure for management of acute stroke due to its invasiveness, the demand for high surgical skills, long operative time, and availability of relatively shorter therapeutic time window. In search of an ideal treatment modality; vascular recanalization strategies gradually evolved over time with the sequential advent of pharmacological thrombolytic therapies, mechanical endovascular devices, and bridge therapy techniques—multimodal reperfusion therapy approach being the latest in this regard.
On arrival in the hospital, a stroke patient undergoes NCCT head to differentiate ischemic stroke from hemorrhagic stroke and stroke mimics, and all the routine investigations are sent simultaneously. Then, patient is immediately shifted to an adjacent MRI suite for stroke protocol MRI (5 minutes)- FLAIR, DWI, and magnetic resonance angiography (MRA). In cases with ischemic stroke, intravenous rt-PA therapy is started if there are no contraindications. Patients with acute large vessel occlusion but small infarct size are immediately shifted for mechanical endovascular therapy (MET). Primary MET is attempted in patients in whom intravenous rt-PA is contraindicated. Cases contraindicated/unlikely to benefit (high clot burden) by MET and cases where MET has failed to recanalize the vasculature are considered the primary and secondary candidates for surgical embolectomy, respectively. In the primary surgical embolectomy cases, SPECT (20 minutes) is done to confirm reduced blood flow in the blocked vessel territory and delineate the relative size of penumbra. In secondary surgical embolectomy (MET failure) cases, repeat-MRI is done to assess the latest infarct size as infarct progression during MET is a contraindication for surgical embolectomy. The surgical candidate (same sized infarct showing mismatch) is shifted to operation room at the earliest possible.
Large infarct size (more than one-third the size of MCA territory) especially those involving lenticulostriate territory.
Under general anaesthesia, patient undergoes frontotemporal craniotomy and sylvian fissure is opened thereby exposing ICA, ACA (A1) and MCA (M1, M2, M3) which are inspected for presence of intraluminal clot. Vessel blocked by intraluminal clot looks bluish-black in colour, firm and slightly expanded non-pulsating segment; and distal pulsations are absent. All the exposed major vessels are inspected along the direction of blood flow to delineate the whole extent of intraluminal clot, which may be extending into multiple branches. To prevent distal migration of intraluminal clot/embolus, distal temporary clips are applied just distal to the vessel point beyond which clot is not seen. Proximal temporary clip is then applied just proximal to the most proximal visible limit of the intraluminal clot. Horizontal arteriotomy (nearer to the distal end of the clot than proximal end of the clot) is made on the non-atherosclerosed portion of clot-filled vessel using Kamiyama microscissors. Portion of clot distal to arteriotomy is removed first by manual pulling of the intraluminal clot through the arteriotomy using bayoneted microforceps, with the assistance of back flow achieved after removing distal temporary clip. As backflow pressure is less forceful, this distal clot portion may be softly and sequentially milked (external pressure) in retrograde direction using bayonet microforceps and pushed out through the arteriotomy. Good backflow confirms vessel patency distal to arteriotomy, following which distal temporary clip is reapplied. A proximal temporary clip is then removed and portion of clot proximal to arteriotomy is removed by manual pulling of the intraluminal clot through the arteriotomy using bayoneted microforceps with the assistance of intraluminal pushing pressure in the gush of proximal blood flow. This intraluminal blood flow pressure forcefully pushes the proximal clot portion out through the arteriotomy, thus recanalizing the vessel proximal to arteriotomy. Proximal temporary clip must be kept ready before completing proximal clot removal for re-application immediately after removing the proximal clot portion, thus avoiding pooling of surgical field with blood. After confirming good forward flow (gush of blood), proximal temporary clip is reapplied immediately. The artery portion which has been emptied off the intraluminal clot is then irrigated with heparin mixed saline to remove the small residual pieces of clot, if any. To reconfirm distal vessel patency, distal temporary clips are removed one by one and patency is confirmed by back flow of blood through arteriotomy. Once recanalization of the previously blocked vasculature is confirmed, proximal and distal temporary clips are now applied just proximal and just distal to arteriotomy, respectively, thereby establishing blood flow in the non-arteriotomized portions of the cerebral vasculature even before arteriotomy closure. Arteriotomy is then closed horizontally using interrupted 8-0 monofilament nylon sutures. Following arteriotomy closure; proximal and distal temporary clips are removed and blood circulation gets re-established in the arteriotomized portion of the cerebral vasculature also, which is confirmed by presence of vascular pulsations and red colour of vessels. Intraoperative arterial Doppler and intraoperative indocyanine green video-angiography are done to re-confirm recanalization of previously blocked large arteries and good anterograde flow in their distal branches, and is regarded as equivalent to TICI 3 flow. During the procedure, systolic blood pressure is elevated till the time of recanalization to enhance collateral blood flow. If any acute re-occlusion (most likely due to residual debris or clot) is identified; arteriotomy just distal to re-occlusion site is reopened, residual or new debris/clot is removed and arteriotomy is re-closed with larger bites in the same manner. Arteriotomy site may be reinforced with a small piece of gelfoam soaked in fibrin glue. After thorough irrigation of subarachnoid space, duramater is closed and bone flap is replaced. In cases with long and/or fragmented intraluminal clot, multiple tandem arteriotomies may be needed to remove the clot completely. In cases with no identifiable intraluminal clot in the intracranial ICA and its major distal branches on inspection, possibility of extracranial ICA occlusion is kept and surgical embolectomy is abandoned. Patient then undergoes STA-MCA bypass in the same sitting for improving vascularization of penumbra. All patients undergo immediate post-operative MRI brain- FLAIR, DWI, and MRA to document postoperative status of infarct and recanalized vessels. All patients receive perioperative antiepileptics, antibiotics, analgesics, and intravenous fluids with respect to their body weight. NCCT head is done at 24 hours after surgery to rule out any intracranial hematoma, following which patient is started on anticoagulation therapy. The retrieved clot is sent for histopathological examination in all cases.
On an average; our preoperative investigation time for primary surgical embolectomy is approximately 40 minutes, incision to reperfusion time is up to 60 minutes, and door to reperfusion time is up to two hours in primary surgical embolectomy cases.
Representative case report [Figure 1], [Figure 2] and [Figure 3]
A 33-years-old lady with past history of cardiac valve replacement few years back (not receiving warfarin) had presented with sudden onset of weakness left half of body. She reached the hospital 4.5 hours after onset and was alert and oriented with left hemiparesis (power 1/5) on admission [NIHSS 9]. Stroke protocol MRI with MRA [Figure 1]A, [Figure 1]B, [Figure 1]C, [Figure 1]D and 99mTc HMPAO SPECT [Figure 2]A showed acute right MCA territory patchy infarct with reduced blood flow due to acute M1 segment occlusion. This case was a contraindication of intra-venous rt-PA because of 3-hour time restriction, but there was good penumbra (DWI-MRA mismatch and DWI-clinical mismatch) that can be rescued by reperfusion. As the Neuro-Cath Lab was not available (another intervention procedure going on), she was immediately taken for surgical embolectomy (6 hours after onset). By right pterional craniotomy and sylvian fissure dissection; right ICA bifurcation, M1 and M2 segments of MCA were exposed. Intraluminal clot was found extending from distal part of pre-bifurcation segment upto post-bifurcation segment of M1. By single transverse arteriotomy, we were able to remove the intraluminal clot completely [Figure 3]A, [Figure 3]B, [Figure 3]C, [Figure 3]D. Immediate postoperative brain MRI with MRA showed good restoration of blood flow in right MCA and its branches, and there was no increase in right MCA territory infarct on diffusion sequence [Figure 1]E, [Figure 1]F, [Figure 1]G, [Figure 1]H. Her incision to reperfusion time was 80 minutes, and door to reperfusion time was 170 minutes. Her symptoms gradually recovered. At 3 months follow-up, she was independent and power in left upper and lower limb were 4/5 and 5/5, respectively.
Acute large vessel occlusion can be atherothrombotic, atheroembolic, atherothromboembolic, cardioembolic, arterial dissection, or idiopathic in origin. Site of clot lodgement depends on its site of origin, size and consistency. Clots with large size and greater stiffness tend to get lodged proximally in the ICA, whereas smaller and softer clots will be able to pass more distally into M1 or M2 part of MCA.
Time is brain. To expedite the treatment, patients with acute ischemic stroke should be subjected to only select imaging studies totally requiring door-to-imaging time of < 20 min and a door-to-imaging interpretation time of < 45 min. Standard neuroimaging protocol in acute stroke includes NCCT head followed by CTA, and MRI brain stroke protocol (DWI, FLAIR ± T2). MRI-DWI being the gold-standard test to measure the “CORE” (acutely infarcted brain tissue) volume as early as 30 min after stroke onset may replace CT as the initial imaging stroke screening tool in future. Perfusion studies (CTP, MRP, SPECT) identify volume of penumbra: severely ischemic but non-infarcted potentially salvageable brain tissue surrounding the non-salvageable infarcted core. Mismatch between various assessing tools (DWI- MRA mismatch, DWI-perfusion mismatch, DWI-clinical mismatch) plays an important role in treatment planning. Though penumbra reduces over time, survival of penumbra up to 48 hours due to good collaterals has been reported in PET-based studies. Salvation of penumbra by timely restoration of its arterial blood supply is the primary therapeutic goal for AIS patients.
Various major randomized trials have reported 58.7–88% successful recanalization rates (TICI 2b and 3) of mechanical stent thrombectomy. Also, as planned studies are usually conducted under ideal circumstances (ideal patients, well trained staff, ideal instruments, etc); it is harder to reproduce similar results in world-wide practice especially in odd hours. Mechanical thrombectomy is likely to be associated with thromboembolic events (distal embolization of clot fragments) despite use of proximal flow control strategies, thus TICI 2 recanalization is more likely than TICI 3 recanalization., For assessing the recanalization status as outcome, majority of the endovascular studies consider both TICI 2 and TICI 3 flows as successful recanalization. The actual complete TICI 3 recanalization is achieved in approximately 40–50% cases only in these studies. The unpreventable distal embolization of clot during endovascular therapy may be the most likely cause of this reduced TICI 3 recanalization, which is unlikely to improve in future despite technological advances.
Better recanalization rates can be achieved with combined pharmacological and mechanical techniques than either of them alone- exception being patients with high clot burden. Hence, intravenous rt-PA within 3–4.5 hours followed by mechanical (endovascular) recanalization within 6–8 hours from time of onset of symptoms have become the gold standard management for acute large vessel occlusion.,, With the recent publication of some extended time window trials (DAWN, DEFUSE 3), endovascular treatment window for AIS may get expanded up-to 24 hours from onset of symptoms.,
As the infrastructure required for mechanical endovascular therapy (MET) is extensive and expensive, it is unrealistic to expect every stroke center to specialize in these neurointerventional techniques. Thus, MET should be carried out only in high volume specialized stroke centers equipped with interventional neuroradiological services round the clock.
The success of a reperfusion therapy solely depends on the time from onset to reperfusion, which has three parts: onset to admission, admission to start therapy, and start therapy to reperfusion. Time from onset to admission is largely influenced by factors outside the hospital; time from admission to start therapy can be reduced to minimum by interdepartmental coordination. The first two parts may remain the same for both the microsurgical and endovascular procedures, it's the last part (start surgery to reperfusion) that makes endovascular procedures (groin puncture to reperfusion: 30–50 minutes) the current standard of care.
Microsurgical embolectomy is an invasive, time-consuming, and high surgical skills demanding complex recanalization procedure to be performed by a specially skilled neurosurgeon under general anaesthesia; and thus could never become a mainstream procedure for acute stroke., However, early temporary clip application before any clot manipulation prevents distal embolic events and provides the highest (90–100%) complete (TICI 3) revascularization rates;, thereby making it an outstanding and may be the gold-standard procedure in the armamentarium of recanalization techniques. Being a high surgical skills demanding procedure, it may be limited to high-volume centers and expert surgeons. It may be more effective than endovascular techniques in achieving complete recanalization of large vascular segments occluded with high clot burdens and involving multiple branch vessels such as ICA terminus occlusions. It can be the “third-line” option for revascularization therapy especially in stroke centers having adequate experience in revascularization surgery procedures.
With the widespread acceptance of endovascular methods as a preferential option to treat acute large vessel occlusion, surgical embolectomy can be preferred in the following situations:,
Some intraoperative technical aspects play important role in its success—(i) clot length- long thrombi extending into multiple branches or peripheral vessels may require multiple arteriotomies thereby delaying reperfusion; (ii) clot consistency- hard thrombi may damage intima during clot removal leading to re-occlusion; (iii) collateral flow- poor collateral flow may decrease back flow which helps flush out the thrombi. All these aspects are equally important in endovascular thrombectomy procedures. Early postoperative anticoagulation (anticoagulation or antiplatelet therapy 24 hours after surgery) after ruling out ICH is recommended to reduce incidence of recurrent embolization., Long-term anticoagulation or antiplatelet therapy is not indicated for surgical embolectomy per se, but may be needed due to the root cause of the thromboembolic event.
Besides the usual surgery- and anaesthesia- related risks; the likely specific complications include arterial leakage through the arteriotomy leading to life-threatening hematoma formation, acute re-occlusion, vasospasm, life-threatening increase in infarct size necessitating decompressive craniectomy and duraplasty, reperfusion injury, hemorrhagic infarction, and delayed vessel stenosis at arteriotomy site.
Though microsurgical embolectomy is the most labor-intensive and cost-intensive of all the 3 (pharmacological thrombolysis, mechanical thrombolysis, and microsurgical embolectomy) currently available options for reperfusion in acute stroke, it is still the most powerful means of completely re-establishing blood flow. Also, it is advantageous over others in terms of (1) low risk of distal embolus migration by appropriate use of temporary clips and antero-/retro-grade blood flow during the procedure, (2) more completeness of revascularization by removing clot in all vessels and branches, and (3) low risk of haemorrhage (sICH) because no thrombolytics/anticoagulants/antiplatelets are used during the procedure. Microsurgical embolectomy is likely to achieve complete recanalization (TICI 3) in nearly 100% cases, whereas the rate of TICI 3 recanalization in the recent endovascular trials remains 40–60%. Also, many of the endovascular procedures declared successful sometimes achieve only TICI 2b (not TICI 3) recanalization. The main disadvantages of microsurgical embolectomy are its steep learning curve, delayed reperfusion, and need of seamless collaboration among various in-hospital services. Also, individual differences in performance may be greater in surgeons than in endovascular neurointerventionists. Multistep multicollaborative surgical embolectomy requires a longer reperfusion time (incision to reperfusion) in comparison to endovascular methods. Thus, the main surgical challenge is to be meticulously quick and achieve complete reperfusion at the earliest possible within the tight therapeutic time window. All these key issues limit the generalizability of surgical embolectomy. However, after sufficient experience, it can achieve complete reperfusion within 60–80 minutes (like Hino et al.); which is nearly comparable to the endovascular thrombectomy procedures. In comparison, MET carries the advantage of being lesser labor-intensive (if not lesser cost-intensive), having shorter reperfusion time than surgery, and not-mandatory need of general anaesthesia.
Like other stroke centers, we prefer endovascular approach as the first-line treatment for acute large vessel occlusions, and reserve surgical embolectomy for endovascular contraindication/failure/hemorrhagic complication/non-availability situations only. We recommend emergency microsurgical embolectomy to be included in the treatment algorithm for management of acute LVO. Hybrid neurovascular facility combining an angiography suite and operating room can reduce the time needed for patient preparation and transport following endovascular treatment failure or complication.
Etminan et al. have recommended a time frame of 6 hours from onset for surgical embolectomy. Park et al. have used 8-hour time window for a surgical embolectomy in their study. However, such fixed time window appears to be highly artificial and unscientific as late (>6 hours after onset) surgical embolectomy may also be beneficial in saving penumbra in patients with good or moderate collateral blood flow.
Microsurgical embolectomy technique can also be useful in management of endovascular complications like irretrievable balloon/coil migration or emboli leading to iatrogenic intracranial vessel occlusion. If microsurgical embolectomy do fails, option of STA- MCA bypass as the next salvage procedure in same sitting is also available.
The major prognostic factors of embolectomy are duration of vessel occlusion, collateral blood flow, site of occlusion, site of origin of embolus, and perioperative management; the first two being the most important. Surgical skills of the operating neurosurgeon and time management skills of the whole team are equally important in determining the final clinical outcome following such rare high-surgical-skills demanding procedures.
Though surgical embolectomy is an invasive, time-consuming, labor-intensive, and high surgical skills demanding procedure; it provides the highest (90–100%) complete (TICI 3) revascularization rates. Thus, microsurgical embolectomy remains an indispensable recanalization tool and should not be dismissed as an antiquated reperfusion method; instead it should be incorporated into guidelines of management of acute large vessel occlusive stroke as a therapeutic procedure, so that it is available as a treatment option of last resort following endovascular contraindication/failure/complication in stroke centers with endovascular expertise available round the clock and as the second treatment option in stroke centers without endovascular expertise. All Vascular Neurosurgeons should develop the required microsurgical skills by attending dedicated cerebrovascular training programs imparting this antique expertise to their trainees. Making appropriate laboratory training models for regular hands-on practice can play a master stroke in imparting and retaining these highly meticulous microsurgical skills.
Because the duration of salvage of ischemic penumbra (depends upon the collateral circulation) varies widely from patient to patient, we recommend more flexibility in giving chance of revascularization to every patient of acute large vessel occlusion ischemic stroke in presence of a salvageable penumbral tissue irrespective of time; instead of following the rigid pre-specified 6–8 hours time window concept.
Financial support and sponsorship
Conflicts of interest
The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
[Figure 1], [Figure 2], [Figure 3]