Joeky Senders

40 Chapter 2 available in the NSQIP registry. This can cause uncontrollable confounding, as is also demonstrated in a previous study. 58 Underestimation in this case might also be caused by underreporting and selection bias as participating hospitals are not obliged to contribute all consecutive cases; however, data is collected on randomly assigned patients by trained surgical reviewers, and inter-rater reliability audits are performed to ensure data reliability. Complications after the 30-day limit used by the database are also unaccounted for in this study. The breadth of multicenter data from NSQIP used in the current study is more representative than most single-center reports; however, the effects of surgeon experience or center volume on postoperative outcomes cannot be accounted for in this database. All surgical studies are limited by variability in surgeon experience, 59 as well as geographical location, 1 both of which have been demonstrated to independently affect complication rates in neurosurgery. Despite these limitations, this study provides useful insight into the rates, reasons, timing, and predictors of major complications, extended length of stay, reoperation, readmission, andmortality after craniotomy for primarymalignant brain tumors. Future studies should focus on building advanced prediction models for short-term outcomes after craniotomy, enabling physicians to tailor postoperative management to the risk profile of the individual patient. A national neurosurgical quality improvement registry including tumor specific and neurosurgical variables can be essential for achieving this goal. Conclusion Among patients undergoing craniotomy for primary malignant brain tumors, 12.9% experienced a major complication within 30 days after surgery, most of which occurred during the initial hospital stay. Intracranial hemorrhage and wound-related complications were the major causes of reoperation and readmission, respectively. ASA-classification and dependent functional status are primarily predictive for morbidity and mortality within 30 days after craniotomy for a primary malignant brain tumor. Future inclusion of tumor and neurosurgical specific variables could allow for a more granular risk assessment of short-term outcomes after craniotomy, but the lack of these variables currently limits the implications of this study.