To the Editor: It was with great pleasure that we read the article by ter Laan et al, "Selective Intensive Care Unit Admission After Adult Supratentorial Tumor Craniotomy: Complications, Length of Stay, and Costs".1 The authors conclude that the implementation of an institutional policy of "no ICU, unless" instead of an "ICU, unless" provides a cost reduction without compromising patient safety. We consider their findings are highly relevant. Indeed, the postoperative neurointensive care of brain tumors prioritizes an early detection and treatment of postoperative complications.2,3 At the same time, reducing costs without compromising the quality of care and patient safety is an area of major global concern. Nevertheless, there are still some discrepancies along their methodology we would like to address. As such, ter Laan et al1 considered the length of stay from 1 d prior to surgery along with the recovery time. Such a consideration may overestimate the total length of stay, reducing the impact on cost reduction and length of stay. Another important issue to note is regarding the influence of a different monitoring on the development of complications in the intensive care unit (ICU) and neurosurgery ward. As authors discussed, adding the fact that the monitoring of patients in the ICU or medium care unit (MCU) has a higher sensibility to detecting complications than that in the neurosurgical ward (due to the lack of highly sensitive procedures), it is a possible overestimation and underestimation, in the ICU and neurosurgery ward, respectively, which might bias the results of this study. Benatar-Haserfaty et al4 reported that only a small fraction of patients require length of stay longer than 1 d after craniotomy (21.5%). The implementation of certain strategies, such as early postoperative computed tomography (EPOCT), is not recommended as a routine use in patients undergoing brain tumor surgeries. Other studies have suggested that a CT scan should be performed only in patients presenting with clinical deterioration.5 Despite that there are some standardized criteria to make the ICU or no-ICU decision more objective, in this study, the subjectivity in the neurosurgeon (based on nonspecified high expected blood loss) or anesthesiologist (based on nonspecified cardiopulmonary comorbidity and functional status) criteria confers caution in interpreting these results. More objective criteria for ICU admission should be regarded in further studies. We agree with the statements introduced in the Aaronson et al6 commentary; however, in clinical practice, we consider it relevant and essential to answer some sort of issues, namely: Is there a need to identify patients who need ICU care and who do not need it? If the patient's care shifted to the neurosurgery ward or step-down ICU, what parameters ought to be monitored? How should the training requirements for the nursing staff be developed? What is the current knowledge of warning signs regarding choosing which place is the best for every patient? In our opinion, further randomized controlled trials are needed to answer these questions. Maybe the "ICU, unless" paradigm would have ended. The answers to the aforementioned questions critically influence the cost and worldwide protocol of the neurosurgical management of patients. Disclosures The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.