In Reply: Necrosectomy Versus Stand-Alone Suboccipital Decompressive Craniectomy for the Management of Space-Occupying Cerebellar Infarctions—A Retrospective Multicenter Study

To the Editor: We would like to express our gratitude to Damodaran et al1 for taking the time to provide their insightful comments on our recently published article titled “Necrosectomy Versus Stand-Alone Suboccipital Decompressive Craniectomy for the Management of Space-Occupying Cerebellar Infarctions—A Retrospective Multicenter Study” in Neurosurgery.2 Below, we present our point-by-point response to the Letter to the Editor. VOLUMETRIC ANALYSIS OF INFARCTED TISSUE Damodaran et al1 accurately pointed out one of our study limitations, namely, the use of computed tomography to evaluate the cerebellar stroke volume. Hypodensities on computed tomography can correspond to both stroke areas and accompanying edema. Incorporation of other imaging modalities, such as magnetic resonance imaging and/or computed tomography perfusion, could help further differentiate between the one and the other and provide more reliable radiological information.3,4 For logistic reasons, we believe that computed tomography perfusion is the more viable method to evaluate the ischemic core as it is faster and more readily available in centers than magnetic resonance imaging. That being said, the latter remains the gold standard for evaluation of the ischemic core.5 While we agree that other imaging modalities would have provided more precise differentiation between edema and ischemic core, our results reflect daily clinical practice, where CT scans are routinely performed in patients with stroke outside of randomized controlled trials and/or prospective studies. EXTENT OF CRANIECTOMY AND RELEVANCE OF NECROSECTOMY As pertinently pointed out by Damodaran et al,1 the extent of craniectomy was not further assessed in our study. For supratentorial decompressive craniectomy, ample research has been conducted to determine the ideal size of the bone flap. The prevailing evidence suggests that the bone flap should measure at least 12 to 14 cm in diameter.6,7 Flaps smaller than this are linked to an increased risk of transcalvarial herniation and edge contusions because of shear forces. Conversely, flaps exceeding 14 cm in diameter are associated with a higher likelihood of bleeding, injury to bridging veins or sinuses, postoperative hydrocephalus, and the sinking skin flap syndrome.8,9 In the posterior fossa, these dimensions have not been studied, and the initial reports dating back to the 1950s presupposed a wide exposure of the cerebellum, from the transverse sinus superiorly to the sigmoid sinus laterally and the foramen magnum caudally10,11; these dimensions are certainly not routinely achieved nowadays. Thus, the extent of bony decompression remains to be validated in a modern cohort. A follow-up multicentric study is underway to evaluate this precise question. Nevertheless, we would like to point out that patients undergoing necrosectomy in this cohort did not receive craniectomies and that their bone flaps were reattached after microsurgical evacuation of the necrotic tissue.12,13 We believe that the removal of avital, necrotic tissue is responsible for the differences observed in functional outcome, for it interrupts the inflammatory responses triggered by the ischemic tissue and thus reduces collateral damage to the surrounding viable parenchyma.14 As the bone flaps were reattached in our cohort, we believe that the extent of bony decompression, while potentially relevant to explain why maybe some patients in the craniectomy group did not achieve a favorable outcome, becomes secondary in the pathomechanisms behind the observed differences in outcome. We further evaluated this hypothesis by conducting a volumetric analysis on the extent of necrotic tissue resection in patients undergoing craniectomies and craniotomies with necrotic tissue resection and found the surgical reduction of infarcted volume to positively affect functional outcome at discharge and 3 months.15 SURGICAL TIMING BIAS We welcome the criticism on the potential bias of not having performed an additional analysis correcting for surgical timing. However, we believe that such analysis would not have yielded significantly different results as it is well known that patients with cerebellar infarction can often have an insidious course.16 In supratentorial ischemic stroke, surgical timing has been amply debated as current evidence suggests that patients operated on at later timepoints can still benefit from surgical intervention.17 While we agree with the comment that an additional analysis might have reduced the probability of bias, we believe that surgical timing alone cannot be a firm determinant of outcome, as our previously published data also suggest.18