Thomas Blauwblomme, Philippe Meyer, and Christian Sainte-Rose
Thirty years after its princeps description, multiple burr hole surgery is yet not recognized as a first-line revascularization procedure in moyamoya angiopathy. Here, we describe the indications, technique, and pitfalls, emphasizing pediatrics patients. Complication rate of this indirect procedure is remarkably low, and efficacy to restore cerebral blood flow, as assessed by imaging and clinical outcome, is at least comparable to other indirect techniques in children.
Keywords: moyamoya angiopathy, multiple burr hole, indirect revascularization
Using cranial burr holes for cerebral revascularization is the consequence of serendipity more than a Cartesian, scientific approach. Spontaneous neovascularization through a burr hole in a child with moyamoya was first observed by Endo et al in 1984.1 A 10-year-old boy with intraventricular hemorrhage was treated with two frontal external ventricular drains. Bilateral encephalo-myo-synangiosis (EMS) was performed 3 months later, and postoperative digital subtraction angiography (DSA) showed marked bilateral neovascularization through the burr holes. This single “burr-hole” technique was therefore performed in five other pediatric cases along with EMS, with excellent clinical and angiographic results.1
Seven years after this princeps publication, Kawaguchi et al published their experience with multiple burr holes, as the sole revascularization supply in a series of 10 adult patients with moyamoya.2 One to four burr holes were drilled on each hemisphere, bilaterally in 8 out of 10 cases. Neovascularization was found in 41/43 burr holes on postoperative angiography performed 3 to 23 months after surgery, along with improvement in cerebral hemodynamics on SPECT studies, and cessation of transient ischemic attacks (TIAs) in 6/6 patients with preoperative ischemic attacks.
Eight years later, results of indirect cerebral revascularization with multiple burr holes were reported in a pediatric series of 14 children.3 The authors increased the number of burr holes to cover the entire cranial vault, through 10 to 24 holes per case. Excellent clinical outcome was observed after surgery, as no child suffered from recurrent ischemic strokes. Postoperative angiography showed good neo vascularization, and the complication rate was low.
Since these pioneer studies, and despite good results, multiple burr holes were rarely reported in the literature,4 and this procedure is more considered as a salvage procedure, or as an adjunct to other direct or indirect techniques.5
There is currently no class A-B evidence in the literature to demonstrate the superiority of direct revascularization on indirect revascularization in pediatric moyamoya. Among indirect revascularization procedures, outcome is comparable between EMS, encephalo-duro-arteriosynangiosis, and burr holes as more than 85% of children are stroke free after surgery.
In our pediatric neurosurgical department, Necker-Enfant Malades in Paris, multiple burr hole surgery is the first-line surgical option for pediatric moyamoya angiopathy. We choose this approach regardless of the underlying etiology, age of the patient, or modality of revelation of the disease.
Although no basic science research has demonstrated how pial anastomosis occur in multiple burr hole surgery, vasculogenesis and angiogenesis are believed to occur at each burr hole because of chronic brain ischemia and vascular growth factor secretion. The principle of the technique is to facilitate the communication between the external (donor) and internal carotid (recipient) arteries systems through bone and meningeal opening. Increasing the number of burr holes increases the surface of brain to be revascularized, in particular the junctional areas (PCA/ MCA or ACA/MCA) in the frontal poles, near the midline and parietoccipital area.
This surgical technique is simple and safe. It does not require transient clipping of the arterial vessels, as needed in bypass surgery, and there is no cosmetic defect associated to temporalis muscle transposition, as with EMS or EDAMS. The complication rate is very low, as we report no permanent neurological deficit or death related to the procedure in our cohort of 64 operated children (transient subcutaneous effusion, n = 5; meningitis, n = 1; superficial skin infection, n= 1).
This approach allows extensive revascularization of the hemispheres. Indeed, burr holes can be placed all over the cranial vault, and revascularization of the frontal pole, occipital lobe, and junctional areas is therefore possible.
Bilateral revascularization is easy during the same procedure, through a unique cosmetic incision, without increasing the morbidity related to the surgery.
This technique does not preclude further revascularization in case of stroke relapse, as the superficial temporal artery (STA) is respected, and other indirect (burr holes, EMS, EDAMS, omentum transposition) and direct techniques (STA- MCA bypass) can be performed despite the initial procedure.
Multiple burr hole surgery is an indirect technique and, as such, revascularization is efficient only a few weeks after the surgery. Therefore, patients with frequent TIAs may undergo ischemic events during the postoperative period. In our cohort of 64 patients, 4 patients had TIAs that resolved spontaneously after a mean delay of 79 days.
As imaging evolves, brain hemodynamics can be studied with magnetic resonance imaging (MRI) techniques. Cerebral blood flow (CBF) can be measured with arterial spin labeling (ASL) MRI,6 along with cerebrovascular reserve with blood oxygen level-dependent (BOLD) MRI, and oxygen extraction fraction. Targeted cerebral reperfusion may therefore be possible, with neuronavigation, to increase CBF where it is needed.
Thickness of the cranial vault can make the surgery challenging in case of sickle cell disease, but is never a formal contraindication. Cerebral atrophy is a concern, as the distance between the brain and the dura makes it difficult for the angiogenesis to occur, and pericerebral fluid collection can be a complication in very atrophic patients.
There is no absolute contra indication to this technique in patients with moyamoya and cerebral hemodynamics impairment.
However, as in all indirect techniques, cortical atrophy is a bad prognosis factor for revascularization, and is associated with an increased risk of complications, like subdural effusion. In case of major atrophy, the balance between risk and benefit shall be carefully weighted.
In order to avoid perioperative complications, preoperative imaging shall be analyzed carefully. Spontaneous transosseous anastomosis need to be localized on the angiography. They must be respected during the subgaleal dissection, the drilling of the burr holes, and the meningeal opening, to avoid perioperative ischemic events.
Delineating areas with cortical atrophy is important, as the burr holes may not be targeted on these areas.
Identification of areas with hemodynamic impairment is crucial, as they are the target of the procedure.
Patients with sickle cell disease deserve special attention among other moyamoya patients. A systemic evaluation is mandatory, as these patients may suffer from cardiac failure, or complications from chronic blood exchanges. Preoperative blood exchange may be necessary, as preoperative hemoglobin levels need to be higher than 10g/dl. Careful analysis of the blood group is required, as these patients may have alloimmunization that needs to be anticipated if perioperative transfusion is required. During surgery, cautious hemostasis of the burr hole is mandatory with bone wax, as the cranial vault may be particularly thick because of the hematopoiesis process.
Avoiding hypoxia and hypothermia is essential to avoid vaso-occlusive crisis.
In our cohort of 64 patients operated with the multiple burr holes techniques, neither mortality nor permanent neurological morbidity was noted.
Possible complications included: transient subdural effusion (n = 5), meningitis (n = 1), and superficial scar infection (n = 1).
Information to the patient must mention that indirect techniques are efficient only a few weeks after the procedure, and therefore ischemic events are possible after the surgery. In our cohort, with a mean follow-up of 270.6 patient years, 89.1% of the patients had no postoperative ischemic events, and a second surgery because of recurrent TIA was required in 3 out of 64 patients.
During anesthesia and recovery, acute and revolving episodes of decreased cerebral perfusion pressure, exposing these patients to a major risk of silent ischemic events, should be strictly avoided. The first imperative is to maintain a stable arterial pressure with a mean arterial pressure level as close as possible from usual measurements in each patient. Continuous invasive arterial pressure monitoring should therefore be initiated during anesthesia and maintained in the immediate postoperative. Excessive vasoplegia with anesthetic drugs should be avoided, and blood losses strictly compensated.
The second imperative is to insure a stable normoventilation during anesthesia and recovery. Since hypocapnia could be a major source of harmfully decreased CBF in these patients, careful continuous monitoring of end tidal carbon dioxide (ETCO2) is critical, and all episodes of hyperventilation should be avoided. A permissive moderate hypoventilation with an ETCO2 in the range of 40 to 45 mm Hg could be a safe goal to attain during mechanical ventilation. In the postoperative period, painful stimulations could generate episodes of relative hyperventilation in children, resulting in decreased arterial CO2, and decreased CBF. Careful staged analgesia should be therefore maintained during recovery, and in the immediate postoperative period.
The patient is positioned supine in a horseshoe headrest, with the head flexed in a neutral position for bilateral approach, or with the head turned on the contralateral side for unilateral procedures.
Subcutaneous infiltration with a saline solution can be used to facilitate the dissection, but is not mandatory.
A bitragial retrocoronal “zigzag” incision is performed for bilateral revascularization, allowing exposure of the entire vault (Fig. 4.1), and a T-shaped incision is done for unilateral approaches (coronal incision and a posterior parietal incision). Section with the monopolar coagulation reduces the blood loss in the pediatric population, and careful hemostasis is required at each step of the procedure.
Subgaleal dissection is performed gently, with particular attention to preserving the subcutaneous vascularization, and notably the STA and its branches. The periosteum is not elevated to preserve the vessels.
The burr holes are made 3 cm apart from each other on the entire exposed vault. Three lines can be determined: 3 cm from the midline (to avoid any bleeding from the bridging veins during the dural opening), above the linea temporalis, and bellow the temporalis muscle.
For each burr hole, the procedure is the same. A 3 cm triangular periosteal flap is elevated, with the tip facing the midline. A burr hole is made with a 1 cm high speed drill. Under the operative microscope, the dura is widely opened in order to avoid cutting the meningeal arterial branches. The arachnoid layer is opened. Cautious hemostasis is obtained with cottonoid patties and gentle irrigation of saline. Cautery is avoided as much as possible to preserve the potential anastomotic vessels. The periosteal flap is then positioned over the brain in the subdural space.
Two-layer watertight closure of the skin and galea is then done with absorbable sutures, and a subcutaneous drainage is left. A compressive head draining is positioned for 5 days, in order to avoid subcutaneous effusion.
No major difficulties may be encountered intraoperatively, as drilling a burr hole is a standard procedure in neurosurgery.
In case of failure of the revascularization procedure, when ischemic events are still reported by the patient more than 3 months after the surgery, further imaging studies are needed. MRI allows the visualization of the transosseous collateral with tetralogy of Fallot studies, and brain hemodynamics can be assessed with perfusion studies (ASL MRI) or BOLD studies. DSA is also necessary to observe if and where pial anastomosis have occurred, and if some burr holes failed to develop collaterals (Fig. 4.2).
A second procedure is then possible, and needs to be targeted to the areas where no collateral developed. Multiple small linear skin incision permits to drill additional burr holes on the ischemic regions, and in our cohort succeeded in two out of two cases to resolve the clinical symptoms. It is important not to reopen the bitragial coronal incision, as it would disrupt some collateral circulation.
If no collateral circulation is noted in any of the burr holes, another revascularization procedure needs to be considered. This was necessary in 1 of the 64 cases in our cohort, and dissection of the STA is possible, and therefore direct (STA-MCA bypass) or indirect procedures are still possible (EDAMS).
The key issue in multiple burr hole surgery is the selection of the patient, as in all direct or indirect revascularization procedure. Careful morphological imaging and hemodynamic studies are necessary to identify the patients who will develop collateral circulations through the burr holes. Indeed, when surgery is offered too late, after ischemic stroke, at the stage of cortical atrophy, functional improvement is minor, and risk of failure and complications is significant. On the other hand, revascularization at a very early stage, without hemodynamic compromise may lead to the impossibility of the brain to develop collateral vessels through the burr holes.
Even if the surgical principle is simple, a meticulous technique is mandatory, and in particular, avoiding excessive coagulation on the dura/arachnoid layers is important.
 Endo M, Kawano N, Miyasika Y, Yada K. Cranial burr hole for revascularization in moyamoya disease.J Neurosurg. 1989; 71(2):180-185
 Kawaguchi T, Fujita S, Hosoda K, et al. Multiple burr-hole operation for adult moyamoya disease.J Neurosurg. 1996; 84(3):468-476
 Sainte-Rose C, Oliveira R, Puget S, et al. Multiple bur hole surgery for the treatment of moyamoya disease in children. J Neurosurg. 2006; 105(6) Suppl:437-443
 Oliveira RS, Amato MCM, Simao GN, et al. Effect of multiple cranial burr hole surgery on prevention of recurrent ischemic attacks in children with moyamoya disease. Neuropediatrics. 2009; 40(6):260-264
 McLaughlin N, Martin NA. Effectiveness of burr holes for indirect revascularization in patients with moyamoya disease-a review of the literature.World Neurosurg. 2014; 81(1):91-98
 Blauwblomme T, Lemaitre H, Naggara O, et al. Cerebral blood flow improvement after indirect revascularization for pediatric moyamoya disease: a statistical analysis of arterial spin-labeling MRI. AJNR Am J Neuroradiol. 2016; 37(4):706-712