In Reply: Superiority of Endovascular Coiling Over Surgical Clipping for Clinical Outcomes at Discharge in Patients With Poor-grade Subarachnoid Hemorrhage: A Registry Study in Japan

To the Editor: We sincerely thank the commenters1 for their professional and thoughtful comments regarding our recent publication in Neurosurgery.2 One concern raised was that patients requiring emergency cerebrospinal fluid (CSF) drainage or decompressive craniectomy (DC) may have been preferentially assigned to the surgical clipping (SC) group. However, these two interventions should be considered separately. In this study, CSF drainage for elevated intracranial pressure (ICP) was presumed to be appropriately implemented according to institutional protocols. We assumed that appropriate ICP management was achieved in both the endovascular coiling (EC) and SC groups. Notably, we did not exclude patients who underwent CSF drainage before EC. However, patients who underwent DC or hematoma evacuation were excluded from the EC group. The efficacy of DC for aneurysmal subarachnoid hemorrhage (aSAH) remains inconclusive and warrants future investigation.3 Our findings do not discourage the use of emergency DC; rather, we believe that patients requiring such interventions should receive appropriate treatment based on individualized clinical assessment. The higher in-hospital mortality rate observed in the EC group in the unmatched analysis suggests that early death related to poor ICP control may be a contributing factor. However, the data were obtained from a high-volume cerebrovascular center where postoperative management is routinely executed with stringent and standardized protocols. Therefore, the higher mortality in the EC group may not be solely attributable to elevated ICP.4 Notably, the EC group included a larger proportion of poor-grade aSAH cases involving the vertebrobasilar artery, which are often managed with EC. Conversely, middle cerebral artery aneurysms, which often require DC, are typically treated with SC.5 To minimize such biases, we conducted a propensity score matching analysis. This enables robust comparisons and supports the generalizability of our findings beyond a narrow clinical subset. Another valid concern is the potential for treatment decision bias due to institutional or physician preferences and experience. Although this is a common limitation in nonrandomized studies, our registry comprises high-volume centers with demonstrated expertise in both SC and EC. These centers follow standardized treatment protocols, conduct multidisciplinary team conferences to formulate treatment strategies, and provide integrated care through the same neurosurgical team. This infrastructure helps minimize institutional bias and supports appropriate treatment selection and implementation. We believe that our study provides valuable insight into the management of poor-grade aSAH (defined as World Federation of Neurosurgical Societies grades III-V) and may serve as a useful indicator to guide treatment strategies in patients with poor-grade aSAH, where the evidence base remains limited.