Joeky Senders

36 Chapter 2 to other types of brain tumors. Additionally, the reoperation rate in glioma patients has not been described in previous multicenter studies. Multicenter studies including brain tumor patients in general demonstrated similar rates of reoperation. 6,13,29 Single-center or single-surgeon studies have reported a wide range of complication rates (4.3-14%) and mortality rates (0-3.7%) after craniotomy for resection of primary malignant brain tumors. 15,30-33 Despite the 30-day timeframe used, the complication and mortality rates determined in this multicenter study (12.9% and 2.6% respectively) fall into the upper ends of the results from observational studies that used a longer follow up. Although these single institutional observational studies provide valuable insight, they can be less generalizable due to their retrospective nature, small sample sizes, and single surgeon or single institutional experience. Because observational studies are often from academic tertiary centers with more specific expertise, the actual rates of postoperative morbidity and mortality may be underestimated in these observational studies. Two studies have previously demonstrated that low-hospital volume is associated with postoperative complications, extended length of stay, and inpatient mortality after surgery for primary brain tumors. 12,14 The high number of participating hospitals in the current study increases the generalizability of our findings; therefore, the results of this study may be more representative of typical management at all hospitals, including but not limited to tertiary care academic centers. Therefore, our results suggest a higher rate of postoperative major complications and mortality after craniotomy for a glioma than previous observational studies have suggested. 15,30-33 To further investigate the generalizability of the current study, the demographics of our patient population were compared with those of the CBTRUS report on primary brain tumors (2009-2013) by Ostrom et al. 1 The age, sex, and race distributions in the current study paralleled the associated distributions stated in the CBTRUS report. These demographic similarities suggest that the study population of the current investigation is representative for primary malignant brain tumor patients across the US. 1 Older age, comorbidity, and dependent functional status have previously been identified as predictors of postoperative complications in glioma patients, 14,34,35 and higher BMI as an additional predictor in brain tumor patients overall. 7 With regards to extended length of stay, older age and comorbidity have been identified as predictors in glioma patients, 12,14 and higher ASA-classification and dependent functional status as additional predictors in brain tumor patients in general. 8 For reoperations, higher BMI has been identified as a predictor in brain tumor patients. 7 For readmission, higher BMI, higher ASA-classification, dependent functional status, and steroid usage have been identified as predictors in brain tumor patients. 4,7,29 For mortality, older age and