Abstract
Objectives
Standard of care treatment for glioblastoma (GBM) involves surgical resection followed by chemoradiotherapy. However, variations in treatment decisions and outcomes exist across hospitals and physicians. In Belgium, where oncological care is dispersed, the impact of hospital volume on GBM outcomes remains unexplored. This nationwide study aims to analyse interhospital variability in 30-day postoperative mortality and 1-/2-year survival for GBM patients.
Methods
Data collected from the Belgian Cancer Registry, identified GBM patients diagnosed between 2016 and 2019. Surgical resection and biopsy cases were identified, and hospital case load was determined. Associations between hospital volume and mortality and survival probabilities were analysed, considering patient characteristics. Statistical analysis included logistic regression for mortality and Cox proportional hazard models for survival.
Results
A total of 2269 GBM patients were identified (1665 underwent resection, 662 underwent only biopsy). Thirty-day mortality rates post-resection/post-biopsy were 5.1%/11.9% (target < 3%/<5%). Rates were higher in elderly patients and those with worse WHO-performance scores. No significant difference was found based on hospital case load. Survival probabilities at 1/2 years were 48.6% and 21.3% post-resection; 22.4% and 8.3% post-biopsy. Hazard ratio for all-cause death for low vs. high volume centres was 1.618 in first 0.7 year post-resection (p < 0.0001) and 1.411 in first 0.8 year post-biopsy (p = 0.0046).
Conclusion
While 30-day postoperative mortality rates were above predefined targets, no association between hospital volume and mortality was found. However, survival probabilities demonstrated benefits from treatment in higher volume centres, particularly in the initial months post-surgery. These variations highlight the need for continuous improvement in neuro-oncological practice and should stimulate reflection on the neuro-oncological care organisation in Belgium.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
In glioblastoma (GBM) treatment, maximal safe surgical resection followed by chemoradiotherapy according to Stupp’s protocol is considered the standard of care [1, 2]. However, treatment decisions, including surgical options, may vary across hospitals and physicians, resulting in treatment variations and possibly outcome variations. If an attempt at complete resection of a GBM results in functional deficits, it not only affects the quality of life but also jeopardizes the chances for timely adjuvant treatment and, consequently, the prognosis [3]. Survival outcomes after surgery for GBM vary across different published reports [4,5,6,7,8,9,10,11,12,13,14,15,16]. Also, among different hospitals within a nation, treatment patterns but also mortality and survival may differ [4, 17,18,19].
So far, correlations between the quality of care or outcome for patients with GBM and the volume of the centre where they are diagnosed, treated and followed-up have not yet been explored in Belgium.
The objective of this nationwide study was to analyse interhospital variability in 30-day postoperative mortality and the 1- and 2- year survival for GBM patients diagnosed between 2016 and 2019 in Belgium. GBM patients who underwent a (stereotactic) biopsy or a surgical resection were investigated separately. Associations between case load of the centre on the one hand, and survival or mortality on the other hand, were explored, taking into account prognostic factors (sex, age, WHO performance status, comorbidities and other tumours).
Materials and methods
Data collection and study cohort
For this study, all GBM patients (ICD-O-3: 9440/3, 9441/3, 9442/3, 9445/3 - C71) diagnosed between 2016 and 2019 were identified in the database of the Belgian Cancer Registry (BCR). (Details in Supplement 1)
Hospital allocation
Benchmarking for mortality and for 1-/2-year survival was based on the centre performing the biopsy or the surgical resection. Patients who underwent a resection after a biopsy were incorporated for postoperative mortality in both groups. Targets for 30-day postoperative mortality were defined by consensus among a multidisciplinary panel, consisting of specialists with known clinical experience and/or previous scientific contributions in neuro-oncology from both academic and non-academic centres, as < 3% after resection and < 5% after biopsy. Both international target reference values and results were considered during this process. The surgical case load of a hospital, enabling mortality and survival analysis in relation to hospital exposure to glioma, was defined as the number of unique patients undergoing at least 1 intervention (biopsy or resection) for a primary or recurrent glioma during the reference period (2016–2019) in the hospital. This means that only 1 intervention per patient per centre was counted and patients operated (biopsy or resection) at different centres were counted once at each centre. The hospital volume (category) was expressed as a total over the 4-year study period.
Statistical analysis for 30-day postoperative mortality
The association between hospital volume and mortality within 30 days after surgical resection or biopsy was investigated using logistic regression models. Details are in Supplement 2.
Statistical analysis for survival analysis
The hazard for all-cause mortality since resection or biopsy was modelled using Cox proportional hazard (PH) models. First the PH assumption was evaluated separately for every case-mix variable and a categorical hospital volume variable. The same case-mix variables as for the 30-day mortality analysis were included. Cumulative hazard curves and comparison of the fitted survival lines from a proportional hazard model to the observed Kaplan-Meier curves were used to assess the PH assumption. Further details in Supplement 3.
Results
Cohort description and centre assignment
A total of 2269 newly diagnosed GBM patients were identified in the BCR database between 2016 and 2019. Surgical resection was performed in 1665 cases, of which 159 received both a biopsy and a surgical resection. A (stereotactic) biopsy was performed in 662 patients (followed or not by a surgical resection). In 99 cases (4.4%), the diagnosis of GBM was based solely on imaging. Surgical procedures and tumour characteristics can be found in Table 1 and Supplement 4. The distribution of the hospitals’ surgical case load is illustrated in Fig. 1.
30-day mortality after resection
At population level, the 30-day postoperative mortality after surgical resection was 5.1% (predefined target < 3%). The 30-day post-resection mortality by case-mix and hospital surgical volume is presented in Supplement 5. A funnel plot illustrates the observed variability among the different centres in Fig. 2A. Although 24 centres had a 30-day post-resection mortality above the predefined target, no centres had a mortality beyond the random variability around the population mean. Results seem comparable in historical comparison periods (2012–2015: 4.4%; 2008–2011: 5.3%) (Supplement 6). (Unadjusted) mortality differs by age group (p = 0.0006) reaching 9.3% for the oldest age group (≥ 75y) and by different WHO-status (p < 0.0001) reaching 22.3% for patients presenting with WHO-score ≥ 3. Significant higher mortality rates were also seen in patients with cardiovascular disease (6.5%; p = 0.0143) or diabetes (10.7%; p = 0.0001). The categorical volume variable for the post-surgery models grouped hospitals with < 57 (low), 58–138 (medium) and > 138 patients (high). Adjusted Odds ratio (OR) for 30-day post-resection mortality by hospital volume groups can be found in Table 2. No statistically significant difference in post-surgical mortality versus surgical case load of the hospital was identified (p = 0.170). Also, the adjusted OR, using hospital volume as a continuous variable, failed to achieve statistical significance (OR = 0.997; 95% CI [0.992, 1.002]; p = 0.3013).
30-day mortality after biopsy
At population level, the 30-day postoperative mortality after biopsy was 11.9%, being above the predefined target (< 5%). The 30-day post-biopsy mortality by case-mix and hospital volume is presented in Supplement 7. A funnel plot illustrates the observed variability among the different centres (Fig. 2B). Although 27 centres had a 30-day post-biopsy mortality above the predefined target, no centres had a mortality beyond the random variability around the population mean. Results were comparable in historical comparison periods (2012–2015: 11.2%; 2008–2011: 9.5%)(Supplement 6). (Unadjusted) mortality differs by age group (p = 0.0134) reaching 17.2% for the oldest age group (≥ 75y) and by different WHO-status (p < 0.0001) reaching 25.4% for patients presenting with WHO-score ≥ 3. Again, significantly higher mortality rates were found in patients with cardiovascular disease (14.7%; p = 0.0190) or diabetes (19.4%; p = 0.0173). The categorical volume variable for the post-biopsy models grouped hospitals with < 57 (low), 57–104 (medium) and > 104 patients (high). Adjusted Odds ratio (OR) for 30-day post-biopsy mortality by hospital volume groups are presented in Table 2. No statistically significant difference in post-biopsy mortality versus surgical case load of the hospital was identified (p = 0.7527). The adjusted OR, using hospital volume as a continuous variable, did not reach statistical significance (OR = 0.996; 95% CI [0.992, 1.000]; p = 0.0579).
Observed 1-/2- year survival probability after resection
The unadjusted 1-year observed survival (OS) for glioblastoma patients after surgical resection was 48.6%, while the unadjusted 2-year OS is 21.3% (Supplement 8). The unadjusted 1-year OS after resection differed by age group (p < 0.0001), dropping to 21.1% for the oldest age group (≥ 75y) and by WHO status at presentation (p < 0.0001), declining to 35.9% for patients with WHO-status 2 and to 29.8% for those with WHO-status ≥ 3. Also, in patients with comorbidities 1-year OS was reduced: diabetes: 33.5%; p < 0.0001; respiratory disease 36.0%; p = 0.0013; cardiovascular disease: 37.2%; p < 0.0001. Median survival time (MST) for GBM after resection was 1 year. The categorical volume variable for the post-surgery models grouped hospitals with < 57, 57–138 and > 138 patients. Kaplan-Meier curves for the unadjusted observed survival stratified by hospital volume groups can be found in Fig. 3A. Differences in survival probabilities for the 3 volume groups remained at 2 years after resection. Higher hazard ratios for all-cause death after resection for low vs. high volume centres (1.618; p < 0.0001) and medium vs. high volume centres (1.426; p = 0.0019) were significant in the first 0.7 year (8.4 months) after resection (Table 3). No statistically significant differences in hazard to die remained between the volume groups in the [0.7y, 2.0y] interval since resection. The optimal binary volume threshold determination resulted in contrasting hospitals with fewer than 120 vs. 120 or more patients for the post-surgery analysis.
Observed 1-/2- year survival probability after biopsy
After biopsy, the unadjusted 1-year OS was 22.4%, while the unadjusted 2-year OS was 8.3% (Supplement 9). The unadjusted 1-year OS after biopsy again varied by age group (p < 0.0001), dropping to 8.3% for the oldest age group (≥ 75y) and by WHO status at presentation (p < 0.0001), decreasing to 13.5% for patients with WHO-status 2 and to 9.0% for those with WHO-status ≥ 3. Similarly, for patients with comorbidities an inferior 1-year survival was found (diabetes: 18.3%; p = 0.0153 and cardiovascular disease: 14.4%; p < 0.0001). After biopsy, MST was 0.4 years. The categorical volume variable for the post-biopsy models grouped hospitals with < 57, 57–104 and > 104 patients. Kaplan-Meier curves for the unadjusted observed survival stratified by hospital volume groups can be found in Fig. 3B. Two years after biopsy, survival probabilities for the 3 volume groups converged. Also post-biopsy, in the first 0.8 year (9.6 months), significant higher hazard ratios for all-cause death for low vs. high volume centres (1.411; p = 0.0046) and medium vs. high volume centres (1.467; p = 0.0016) were identified (Table 3). However, in the time period [0.8y, 2.0y] after biopsy, the hazard ratios for the low and medium centres versus the high-volume centres become smaller than 1, which is significant for the medium volume group only (0.527; p = 0.0039). The optimal binary volume threshold determination resulted in contrasting hospitals with fewer than 100 vs. 100 or more for the post-biopsy analysis. Kaplan Meier curves for optimized binary volumes can be found in Supplement 10.
Discussion
The 30-day mortality at population level was 11.9% after biopsy and 5.1% after surgical resection, being above the predefined targets (5% and 3% respectively). Higher rates for 30-day mortality after biopsy and resection were found in elderly patients (≥ 75y) (17.2% and 9.3%) and in patients with worse WHO-status (e.g. WHO-status ≥ 3: 25.4% and 22.3%). International comparison is difficult since most series consider postoperative mortality for both resection and biopsy combined or evaluate different tumour entities, sometimes including meningioma, together. In Spain (2008–2010) for glioblastoma, the 30-day mortality after resection (4.4%) was comparable with the result in this series, but 30-day mortality after biopsy (14.2%) is slightly higher [9]. In Scotland (2021) the 30-day mortality rate for brain/CNS cancer was 5.0% at population level, but figures differ regionally from 2.3 to 9.0%, while in the Netherlands (only GBM) it was 5.2% [4, 20]. In the Danish Neuro-Oncology Registry, post-surgical mortality for all types of glioma ranged from 2% to 4% between 2010 and 2014 (target set at < 5%) [21]. In the US, based on the Nationwide Inpatient Sample, in hospital mortality rate was 2.3% for needle biopsy, 3.2% for lobectomy and 2.7% for other resections [22]. In single centre series in different countries 30-day mortality rates between 1.5% and 10.7% were reported [9, 21, 23,24,25,26,27,28,29,30,31,32,33,34]. De Witt Hamer concluded that mortality within 30 days is not a useful quality indicator for glioblastoma-related complications. This is probably true if biopsy and resection are combined. If only resection is considered, early mortality in patients with irresectable tumours in whom only a biopsy is taken, will not confound post-resection operative mortality rates [4]. Indeed, mortality after biopsy accounts for both fatal procedural (haemorrhagic) complications and for true rapid disease progression. However, reporting post-biopsy mortality for glioblastoma patients could also reveal aberrant rates indicating poor surgical technique or poor case selection and futile tissue sampling. Anyway, the high mortality rates after biopsy in elderly patients and in those with worse WHO-scores (sometimes presenting with a pathognomonic MRI) highlight the need for careful multidisciplinary consideration during medical decision making in these subgroups. In this series, 99 patients (4.4% of all registered GBM cases) were withhold from tissue sampling and received presumably supportive care after radiological diagnosis. No differences in early mortality (after biopsy and resection) were identified between the 3 volume groups in this study, contrasting the findings in the Netherlands by De Witt Hamer, who demonstrated an association of higher case volume with lower early mortality [4]. In that series, an estimated boundary between higher and lower than average early mortality was found at 45 patients/year [4]. Also, in a retrospective analysis of a population based cohort in the state of New York (2005–2014), not only the 30-day mortality rate after brain tumour resection (not only glioma) was lower when surgery was performed by surgeons with a high annual (n ≥ 24) and high cumulative volume (≥ 114 in 5 years), but also a lower rate of complications and reoperations within 30 days [35]. Contrarily, in Spain, no relationship between the number of major complications (including 30-day mortality) and case load of the surgical centres was identified [9]. Similarly, in England, with a larger median hospital case load (n = 273 over a 3-year period), suggesting greater centralisation, the 30-day mortality of intracranial neoplasm resection was not related to hospital volume, but to the individual a surgeon’s case load. A doubling of a surgeon’s caseload resulted in a 20% risk reduction. This suggests a need for further sub-specialisation among neurosurgeons, even within already well-concentrated brain surgery care [36].
International comparison of survival at population level for the subgroups of biopsied or resected patients is cumbersome, since most series don’t differentiate survival results. The 1-year observed survival probability since surgical resection (48.6%) is comparable with the result in Denmark after partial resection (est. 50% on Kaplan-Meier plot), but inferior to the result for the group in Denmark after complete resection (est. 70% on Kaplan-Meier plot) [37].
In Western Norway, MST after resection was 13.7 months and after biopsy 8.3 months, both being superior to the results in our cohort (8.3 months and 4.8 months resp.) [38]. The 1-year survival after biopsy (22.4%) was comparable to the result in Denmark (est. 20% on Kaplan-Meier plot), but superior to the Netherlands were 1-year survival after biopsy and chemoradiation was 15% [37, 39]. In Italy, 1-year survival probability after biopsy for the glioblastoma patients being treated with radiotherapy, was superior to the survival probability for all biopsied patients in Belgium (36.8% vs. 22.4%) while the 2-year probabilities were comparable (7.6% vs. 8.3%) [13]. Two-year post-surgical survival (21.3%) was only slightly inferior to the 2-year survival of 26.5% in the Stupp trial, where 84% of patients underwent resection [1].
Taking into account case-mix variability, survival probability was worse in the first 0.7 years after surgical resection (0.8 years after biopsy) in low- (OR: 1.883; p < 0.0001) and medium- (1.419; p = 0.0019) vs. high-volume centres. This is again in contrast to the findings in the Netherlands, where higher volume was not related to 2-year survival, but this mixed cohort consisted of both biopsied and resected patients [4]. In Finland, all intracranial surgeries are performed in five university hospitals. Patients with newly diagnosed GBM had a 19% higher relative excess risk (RER) when diagnosed and treated in low-volume centre (mean annual case load = 17) than in the high- and medium-volume hospitals (mean annual case load = 54 and 40 resp.) [19]. The optimal binary volume threshold in this series determined at 120 patients/4 years for resection and 100/4 years for biopsy, suggests that GBM patients might have a better outcome when undergoing surgical treatment in a centre performing on average at least 30 glioma surgeries annually.
In a recent survey by the EANS (European Association of Neurosurgical Societies) the majority of respondents agreed that an expert reaches an annual case load of more than 50 consisting of gliomas, meningiomas and metastases [40]. In Belgium, ideas about organisation and centralisation of patients with rare and complex tumours were launched based on studies by the Belgian Health Care Knowledge Centre (KCE) [41, 42]. According to the definition of RARECARE, rare cancers have an incidence rate less than 6/100.000 persons/year, categorising almost all subtypes of primary brain tumours (including gliomas) as rare [43]. Grade 4 glial tumours, mainly molecular glioblastoma (age-standardized incidence rate (ESR) = 6.72), are just surpassing this threshold [44]. Complex cancers are those in specific and difficult anatomical regions and those requiring high technological or expensive infrastructure [41]. Although subjective, the infrastructure and technologies required for state-of-the-art neuro-oncological surgeries (e.g., neuronavigation, neuro-monitoring, awake brain mapping, fluorescence-guided resection,…) could meet this requirement. These findings and considerations seem to contrast with the current dispersion with 53 centres providing surgical care for GBM (Fig. 1). (In 2019, Belgium had 103 hospitals, 7 of which were university hospitals [45]). The evaluation of the use of the aforementioned technologies and the search for an association with mortality and/or survival in GBM was beyond the scope of this study. Previously, these surgical aids were not considered standalone quality (structure) indicators by our group, but rather tools in achieving maximal safe resection [46].
This population-based study shows for the first time an association of hospital volume and survival after resection/biopsy in patients with a glioblastoma. While in this study expertise is defined by a centre’s surgical case load, it is also crucial to recognize that the management of GBM is multidisciplinary, involving neuropathologists, neuroradiologists, radiation oncologists, medical oncologists, and paramedical staff. As a result, centres with higher surgical load typically also handle a higher volume of cases in the other involved specialties, fostering a group expertise, which might contribute to the improved outcome observed in these centres. This is supported by the work of Haque et al. showing that patients with histologically confirmed GBM receiving chemoradiation at high volume facilities have an improved survival [47]. Moreover, decisions to withhold GBM patients from surgery, adjuvant treatment or, any treatment at all, should ideally be made by a multidisciplinary expert team, of course in consent with the patients and their proxies.
This study also harbours limitations. First, outcome indicators were evaluated based on the WHO-2016 classification, still incorporating the prognostically more favourable IDH-mutant glioblastoma (now grade 4 IDH-mutant astrocytoma). However, the limited number of cases (n = 16; 0.5%) likely had minimal impact on the survival. Survival and incidence for molecularly defined glial tumours have been previously published by our group, demonstrating a MST for IDH-wildtype glioblastoma of 9.3 months [44]. Another limitation is that true postoperative mortality cannot be distinguished from mortality due to rapid disease progression, as the exact cause of death is unknown in this cohort. Next, a minor limitation is the exclusion of patients who died at incidence date, potentially following a resection or biopsy (n = 2). However, in these rare instances, it is challenging to determine whether death resulted from (surgical) complications or was attributable to the possible moribund status upon presentation of a (probably haemorrhagic) GBM. Finally, possible variations in upfront excluding poor-prognosis patients with a radiological suspicion for GBM (4.4% in this series) from a diagnostic biopsy or a resection might influence final outcome data in the pathologically proven GBM populations.
Conclusion
Reporting postoperative mortality rates and survival for GBM patients is crucial for understanding outcome, at the national and hospital level. High mortality rates after biopsy in elderly patients and in patients with worse WHO-scores highlight the need for careful consideration during medical decision making in these subgroups. Although in this nationwide analysis no correlation between hospital volume and 30-day postoperative mortality for both biopsied and resected GBM patients could be detected, the overall mortality rates being above the predefined targets, should prompt neuro-oncological surgeons to continuously improve their practice and consider carefully therapeutic/diagnostic procedures in elderly and frail patients. Both 1- and 2-year post-biopsy and post-resection survival are in line with international findings. However, significant differences related to hospital case load should open reflections on the organisation of neuro-oncological care in Belgium.
Data availability
The cancer cohort data used and analysed during the study are available via the Belgian Cancer Registry (BCR) upon reasonable request. The pseudonymized data can be provided within the secured environment of the BCR after having been guaranteed that the applicable GDPR regulations are applied.
References
Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ et al (2005) Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352(10):987–996
Stupp R, Hegi ME, Mason WP, van den Bent MJ, Taphoorn MJ, Janzer RC et al (2009) Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial. Lancet Oncol 10(5):459–466
McGirt MJ, Mukherjee D, Chaichana KL, Than KD, Weingart JD, Quinones-Hinojosa A (2009) Association of surgically acquired motor and language deficits on overall survival after resection of glioblastoma multiforme. Neurosurgery 65(3):463–469 discussion 9–70
De Witt Hamer PC, Ho VKY, Zwinderman AH, Ackermans L, Ardon H, Boomstra S et al (2019) Between-hospital variation in mortality and survival after glioblastoma surgery in the Dutch Quality Registry for Neuro surgery. J Neurooncol 144(2):313–323
Sanai N, Polley MY, McDermott MW, Parsa AT, Berger MS (2011) An extent of resection threshold for newly diagnosed glioblastomas. J Neurosurg 115(1):3–8
Li YM, Suki D, Hess K, Sawaya R (2016) The influence of maximum safe resection of glioblastoma on survival in 1229 patients: can we do better than gross-total resection? J Neurosurg 124(4):977–988
Pan IW, Ferguson SD, Lam S (2015) Patient and treatment factors associated with survival among adult glioblastoma patients: a USA population-based study from 2000–2010. J Clin Neurosci 22(10):1575–1581
Woehrer A, Bauchet L, Barnholtz-Sloan JS (2014) Glioblastoma survival: has it improved? Evidence from population-based studies. Curr Opin Neurol 27(6):666–674
Graus F, Bruna J, Pardo J, Escudero D, Vilas D, Barcelo I et al (2013) Patterns of care and outcome for patients with glioblastoma diagnosed during 2008–2010 in Spain. Neuro Oncol 15(6):797–805
Gulati S, Jakola AS, Johannesen TB, Solheim O (2012) Survival and treatment patterns of glioblastoma in the elderly: a population-based study. World Neurosurg 78(5):518–526
Yabroff KR, Harlan L, Zeruto C, Abrams J, Mann B (2012) Patterns of care and survival for patients with glioblastoma multiforme diagnosed during 2006. Neuro Oncol 14(3):351–359
Mathiesen T, Peredo I, Lönn S (2011) Two-year survival of low-grade and high-grade glioma patients using data from the Swedish Cancer Registry. Acta Neurochir (Wien) 153(3):467–471
Scoccianti S, Magrini SM, Ricardi U, Detti B, Buglione M, Sotti G et al (2010) Patterns of care and survival in a retrospective analysis of 1059 patients with glioblastoma multiforme treated between 2002 and 2007: a multicenter study by the Central Nervous System Study Group of Airo (Italian Association of Radiation Oncology). Neurosurgery 67(2):446–458
Bauchet L, Mathieu-Daude H, Fabbro-Peray P, Rigau V, Fabbro M, Chinot O et al (2010) Oncological patterns of care and outcome for 952 patients with newly diagnosed glioblastoma in 2004. Neuro Oncol 12(7):725–735
Darefsky AS, King JT Jr., Dubrow R (2012) Adult glioblastoma multiforme survival in the temozolomide era: a population-based analysis of Surveillance, Epidemiology, and end results registries. Cancer 118(8):2163–2172
Arrigo RT, Boakye M, Skirboll SL (2012) Patterns of care and survival for glioblastoma patients in the veterans population. J Neurooncol 106(3):627–635
Zhu P, Du XL, Zhu JJ, Esquenazi Y (2019) Improved survival of glioblastoma patients treated at academic and high-volume facilities: a hospital-based study from the National Cancer Database. J Neurosurg 132(2):491–502
Chang SM, Parney IF, Huang W, Anderson FA Jr., Asher AL, Bernstein M et al (2005) Patterns of care for adults with newly diagnosed malignant glioma. JAMA 293(5):557–564
Raj R, Seppa K, Luostarinen T, Malila N, Seppala M, Pitkaniemi J et al (2020) Disparities in glioblastoma survival by case volume: a nationwide observational study. J Neurooncol 147(2):361–370
Liaquat I, Campbell L, McMahon J (2021) Audit Report 2021 on Brain and Central Nervous System Cancers: Quality Performance Indicators.
Hansen S (2016) The Danish Neuro-Oncology Registry. Clin Epidemiol 8:629–632
Barker FG 2nd, Curry WT Jr., Carter BS (2005) Surgery for primary supratentorial brain tumors in the United States, 1988 to 2000: the effect of provider caseload and centralization of care. Neuro Oncol 7(1):49–63
Chang SM, Parney IF, McDermott M, Barker FG 2nd, Schmidt MH, Huang W et al (2003) Perioperative complications and neurological outcomes of first and second craniotomies among patients enrolled in the Glioma Outcome Project. J Neurosurg 98(6):1175–1181
Lofgren D, Valachis A, Olivecrona M (2022) Risk for morbidity and mortality after neurosurgery in older patients with high grade gliomas - a retrospective population based study. BMC Geriatr 22(1):805
Sastry RA, Pertsch NJ, Tang O, Shao B, Toms SA, Weil RJ (2020) Frailty and outcomes after craniotomy for brain tumor. J Clin Neurosci 81:95–100
Solheim O, Jakola AS, Gulati S, Johannesen TB (2012) Incidence and causes of perioperative mortality after primary surgery for intracranial tumors: a national, population-based study. J Neurosurg 116(4):825–834
Stummer W, Koch R, Valle RD, Roberts DW, Sanai N, Kalkanis S et al (2019) Intraoperative fluorescence diagnosis in the brain: a systematic review and suggestions for future standards on reporting diagnostic accuracy and clinical utility. Acta Neurochir (Wien) 161(10):2083–2098
Long D, Gordon T, Bowman H, Etzel A, Burleyson G, Betchen S et al (2003) Outcome and cost of craniotomy performed to treat tumors in regional academic referral centers. Neurosurgery 52(5):1056–1063 discussion 63– 5
Sawaya R, Hammoud M, Schoppa D, Hess KR, Wu SZ, Shi WM et al (1998) Neurosurgical outcomes in a modern series of 400 craniotomies for treatment of parenchymal tumors. Neurosurgery 42(5):1044–1055 discussion 55– 6
Cardona MTS, Fabre ER, Bandera MP, Dominguez W, Ramos TC (2014) Mortality from malignant brain tumors within the first 30 days after surgery. Revista Cubana De Cirurgia 53(3):244–255
Rabadan AT, Hernandez D, Eleta M, Pietrani M, Baccanelli M, Christiansen S et al (2007) Factors related to surgical complications and their impact on the functional status in 236 open surgeries for malignant tumors in a latino-american hospital. Surg Neurol 68(4):412–420 discussion 20
Lassen B, Helseth E, Ronning P, Scheie D, Johannesen TB, Maehlen J et al (2011) Surgical mortality at 30 days and complications leading to recraniotomy in 2630 consecutive craniotomies for intracranial tumors. Neurosurgery 68(5):1259–1268 discussion 68– 9
Liaquat I, Campbell L, McMahon J (2022) Audit report: Brain and Central Nervous System Cancers Quality Performance indicators. Report of the 2020 clinical Audit Data. Scottish Adult Neuro Oncology Network, p 41
Trinh VT, Davies JM, Berger MS (2015) Surgery for primary supratentorial brain tumors in the United States, 2000–2009: effect of provider and hospital caseload on complication rates. J Neurosurg 122(2):280–296
Ramakrishna R, Hsu WC, Mao J, Sedrakyan A (2018) Surgeon Annual and cumulative volumes predict early postoperative outcomes after Brain Tumor Resection. World Neurosurg 114:e254–e66
Williams M, Treasure P, Greenberg D, Brodbelt A, Collins P (2016) Surgeon volume and 30 day mortality for brain tumours in England. Br J Cancer 115(11):1379–1382
Hansen S, Rasmussen BK, Laursen RJ, Kosteljanetz M, Schultz H, Norgard BM et al (2018) Treatment and survival of glioblastoma patients in Denmark: the Danish Neuro-Oncology Registry 2009–2014. J Neurooncol 139(2):479–489
Bjorland LS, Fluge O, Gilje B, Mahesparan R, Farbu E (2021) Treatment approach and survival from glioblastoma: results from a population-based retrospective cohort study from Western Norway. BMJ Open 11(3):e043208
Ho VK, Reijneveld JC, Enting RH, Bienfait HP, Robe P, Baumert BG et al (2014) Changing incidence and improved survival of gliomas. Eur J Cancer 50(13):2309–2318
Gousias K, Hoyer AP, Mazurczyk LA, Bartek J Jr., Bruneau M, Ciltikci E et al (2024) Expertise in Surgical Neuro-oncology. Results of a survey by the EANS Neuro-Oncology section. Brain Spine. 4:102822
Stordeur S, Vrijens F, Henau K, Schillemans V, De Gendt C, Leroy R (2014) Organisatie Van De zorg voor volwassenen met een zeldzame of complexe kanker– synthese. In: (HSR) HSR, editor. Belgian Health Care Knowledge Center (KCE), Brussel
Belgian Health Care Knowledge Center (KCE) (2014)
Gatta G, van der Zwan JM, Casali PG, Siesling S, Dei Tos AP, Kunkler I et al (2011) Rare cancers are not so rare: the rare cancer burden in Europe. Eur J Cancer 47(17):2493–2511
Pinson H, Silversmit G, Vanhauwaert D, Vanschoenbeek K, Okito JK, De Vleeschouwer S et al (2023) Epidemiology and survival of adult-type diffuse glioma in Belgium during the molecular era. Neuro Oncol. 26(1):191–202
Blikvanger gezondheidszorg Algemene Ziekenhuizen. Brussels: Federal Public Service Health, Food Chain Safety and Environment; 2019 https://www.health.belgium.be/sites/default/files/uploads/fields/fpshealth_theme_file/blikvanger_gezondheidszorg_az_v10.pdf
Vanhauwaert D, Pinson H, Sweldens C, Du Four S, Van Eycken L, consortium Q et al (2022) Quality indicators in neuro-oncology: review of the literature and development of a new quality indicator set for glioma care through a two-round Delphi survey. J Neurooncol. (2):365–376
Haque W, Verma V, Butler EB, Teh BS (2017) Definitive chemoradiation at high volume facilities is associated with improved survival in glioblastoma. J Neurooncol 135(1):173–181
Acknowledgements
We would like to thank the following experts that were involved in the validation of the methodology of the technical documentation sheets prior to definite calculation: Frank Weyns, Ludo Vanopdenbosch, Alex Michotte, Nick Liefhooghe, Martin Lammens, Arnaud Lombard, Nicolas Whenham, Ann Tieleman, Siska Dedeurwaerdere, Karolien Goffin. We are very grateful to the following colleagues for their help and support: Jeroen Van Lerbeirghe, Stephanie Du Four, Wim Maenhoudt and Olivier Van Damme. We also would like to thank the contribution of all employees of the Belgian Cancer Registry (BCR) who participated in data collection and processing.
Funding
This research was financially supported by Stichting tegen Kanker, AZ Delta vzw and Ghent University Arne Lannoy, A.K.A. Zorro Fund.
Author information
Authors and Affiliations
Contributions
DV: Conceptualization; Data curation; Formal analysis; Project administration; Writing - original draft; Writing - review & editing. GS, GC: Data curation, Formal analysis; Methodology, Visualization, Writing - review & editing. KV: Data curation; Formal analysis; Methodology; Project administration; Visualization; Writing - review & editing. DD: Validation; Writing - review & editingPC: Validation; Writing - review & editing. RS: Validation; Writing - review & editing. SDV: Supervision; Validation; Writing - review & editing. TB: Supervision; Validation; Writing - review & editing. CDG: Methodology, Supervision; Validation; Project administration; Writing - review & editing.
Corresponding author
Ethics declarations
Competing interests
SDV is a certified Gliolan (Medac GmbH) Trainer and has performed consultancy for Lamepro NV (Gliolan) for which a fee was received. PC has performed consultancy for Bayer, BMS, Leo pharma, Merck, MSD, Rakuten Medical and Servier, and serves as a member of the CTG/CRM in Belgium. All other authors have no relevant financial or non-financial interests to disclose.
Statement of ethics
This study was approved by the Ethics Committee of Ghent University Hospital (DA 2020-042). Written informed consent from participants was not required in accordance with local/national guidelines.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Vanhauwaert, D., Silversmit, G., Vanschoenbeek, K. et al. Association of hospital volume with survival but not with postoperative mortality in glioblastoma patients in Belgium. J Neurooncol (2024). https://doi.org/10.1007/s11060-024-04776-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11060-024-04776-2