Neuro-Oncology Variation in postoperative outcomes of patients with intracranial tumors: insights from a prospective international cohort study during the COVID-19 pandemic

Abstract

Background. This study assessed the international variation in surgical neuro-oncology practice and 30-day outcomes of patients who had surgery for an intracranial tumor during the COVID-19 pandemic. Methods. We prospectively included adults aged ≥18 years who underwent surgery for a malignant or benign in-tracranial tumor across 55 international hospitals from 26 countries. Each participating hospital recorded cases for 3 consecutive months from the start of the pandemic. We categorized patients' location by World Bank income groups (high [HIC], upper-middle [UMIC], and low-and lower-middle [LLMIC]). Main outcomes were a change from routine management, SARS-CoV-2 infection, and 30-day mortality. We used a Bayesian multilevel logistic regression stratified by hospitals and adjusted for key confounders to estimate the association between income groups and mortality.

Full text

Neuro-Oncology
XX(XX), 1–11, 2023 | https://doi.org/10.1093/neuonc/noad019 | Advance Access date 13 April 2023
1
© The Author(s) 2023. Published by Oxford University Press on behalf of the Society for Neuro-Oncology.
Variation in postoperative outcomes of patients with
intracranial tumors: insights from a prospective
international cohort study during the COVID-19
pandemic
Michael T.C.Poon , Rory J.Piper, NqobileThango, Daniel M.Fountain, Hani J.Marcus ,
LauraLippa, FrancoServadei, Ignatius N.Esene, Christian F.Freyschlag, Iuri S.Neville,
GailRosseau, KarlSchaller, Andreas K.Demetriades, Faith C.Robertson, Peter J.Hutchinson,
Stephen J.Price, Ronnie E.Baticulon, James C.Glasbey, AneelBhangu, Michael D.Jenkinson,
Angelos G.Kolias; from the Writing Group of the COVIDSurg-Cancer neurosurgery investigators
on behalf of the COVIDSurg Collaborative; British Neurosurgical Trainee Research Collaborative;
WFNS Young Neurosurgeons Committee; NIHR Global Health Research Group on Acquired Brain
and Spine Injury†
All author afliations are listed at the end of the article
†Collaborating authors are listed below Acknowledgments section
Corresponding Authors: Michael TC Poon, Department of Clinical Neurosciences, Royal Infirmary of Edinburgh,
50 Little France Crescent, Little France, EH16 4SA, UK (michael.poon@ed.ac.uk); Angelos G Kolias, Division of
Neurosurgery, Department of Clinical Neurosciences, Addenbrooke’s Hospital and University of Cambridge,
Cambridge Biomedical Campus, Cambridge CB20QQ, UK (ak721@cam.ac.uk).
Abstract
Background. This study assessed the international variation in surgical neuro-oncology practice and 30-day out-
comes of patients who had surgery for an intracranial tumor during the COVID-19 pandemic.
Methods. We prospectively included adults aged ≥18 years who underwent surgery for a malignant or benign in-
tracranial tumor across 55 international hospitals from 26 countries. Each participating hospital recorded cases for 3
consecutive months from the start of the pandemic. We categorized patients’ location by World Bank income groups
(high [HIC], upper-middle [UMIC], and low- and lower-middle [LLMIC]). Main outcomes were a change from routine
management, SARS-CoV-2 infection, and 30-day mortality. We used a Bayesian multilevel logistic regression strati-
fied by hospitals and adjusted for key confounders to estimate the association between income groups and mortality.
Results. Among 1016 patients, the number of patients in each income group was 765 (75.3%) in HIC, 142 (14.0%)
in UMIC, and 109 (10.7%) in LLMIC. The management of 200 (19.8%) patients changed from usual care, most com-
monly delayed surgery. Within 30 days after surgery, 14 (1.4%) patients had a COVID-19 diagnosis and 39 (3.8%)
patients died. In the multivariable model, LLMIC was associated with increased mortality (odds ratio 2.83, 95%
credible interval 1.37–5.74) compared to HIC.
Conclusions. The first wave of the pandemic had a significant impact on surgical decision-making. While the in-
cidence of SARS-CoV-2 infection within 30 days after surgery was low, there was a disparity in mortality between
countries and this warrants further examination to identify any modifiable factors.
Key Points
•COVID-19 research collaborative efforts allowed international comparisons.
•Low- and low-middle-income countries were associated with higher 30-day mortality.
•This disparity required clarification and identification of modifiable factors.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License
(https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any
medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

2 Poon et al.: Variation in postoperative neuro-oncology outcomes
The impact of the COVID-19 pandemic on the delivery of
surgical care is substantial. It has been estimated that over
28.4 million elective operations were canceled or delayed
during the 12-week first wave of the pandemic worldwide.1
International studies have demonstrated that preopera-
tive SARS-CoV-2 infection is associated with a higher risk
of 30-day postoperative mortality.2,3 Longer-term direct
and indirect effects of the pandemic are yet to be realized
though these are likely to result in excess mortality among
people with cancer due to delays in diagnosis and treat-
ment.4 Few studies have evaluated neuro-oncology serv-
ices during the pandemic5–7 but they do not provide a
global view. Estimating the effect of the pandemic on the
initial management of brain tumors during the first wave
(January–August 2020) can set a reference to compare
hospital activities as the pandemic evolves and new evi-
dence emerges.
Even before the COVID-19 pandemic, the landscape
of global neurosurgery was disparate. A major deficit of
neurosurgeons predominantly in low- and middle-income
countries has resulted in an estimated 5 million essential
neurosurgical procedures not performed each year.8 While
the direct impact of access to neurosurgery cannot be
measured for neuro-oncology patients worldwide, it is rea-
sonable to assess the variations in neuro-oncology prac-
tices of different countries since healthcare systems and
patient pathways can affect patient outcomes.
The aim of this study was to assess the changes to rou-
tine neuro-oncology management that resulted from the
COVID-19 pandemic, and to compare 30-day postoperative
mortality between countries of different income groups.
Methods
Study Design
The COVIDSurg-Cancer is an international, observa-
tional cohort study that assessed treatment pathways
and perioperative events in patients undergoing surgery
for a tumor during the pandemic.9 This study also pre-
sents a unique opportunity to assess patient presenting
features, neuro-oncology practice, and short-term sur-
gical outcomes in different countries. Investigators
from participating centers obtained the appropriate
study approval according to the local and national
requirements.
This study was a preplanned subgroup analysis of pa-
tients from the COVIDSurg-Cancer study who underwent
surgery for an intracranial tumor during the first wave
(January–August 2020) of the pandemic. Any hospital pro-
viding brain tumor surgery in an area affected by COVID-19
was eligible and participation was voluntary. Each inves-
tigator identified a start date for the respective center.
This start date corresponded to the date of admission of
the first patient with confirmed SARS-CoV-2 infection in
the hospital. In hospitals operating a Covid-free surgical
pathway, the start date was the date of admission of the
first SARS-CoV-2 positive patients in another hospital in
the city. Patient recruitment ended 3 months after the start
date. The follow-up period was 30 days after tumor sur-
gery. Collaborators entered anonymized data into a secure
server using the Research Electronic Data Capture online
system.1
Participants
Collaborators recruited all consecutive adults aged ≥18
years who underwent any surgery for an intracranial tumor
during the 3-month recruitment period. Patients with pri-
mary or secondary malignant or nonmalignant tumors
were eligible. Collaborators reviewed hospital records to
collect information about postoperative outcomes.
Definitions of Co-variables
The explanatory variable of interest was The World Bank
income group 2020 (https://data.worldbank.org/country)
of the country where each participating hospital was lo-
cated. The main outcome of interest was all-cause mor-
tality within 30 days of tumor surgery. Collaborators
ascertained mortality data based on their hospital records.
We collected baseline, operative, and tumor character-
istics of included patients. Healthcare system character-
istics included the local 14-day SARS-CoV-2 cumulative
notification rate, COVID-19 free surgical pathway, preoper-
ative Covid screening, and preoperative swab test results.
Community SARS-CoV-2 incidence is a proxy measure of
the risk of SARS-CoV-2 infection and was calculated for
Importance of the Study
Globally there is a major decit of neurosurgeons
predominantly in low- and low middle-income coun-
tries (LLMIC). There is a general paucity of studies re-
porting postoperative outcomes in LLMIC. COVID-19
collaborative surgical research provided an opportu-
nity to assess neuro-oncology practice and outcomes
across countries. Data from our prospective interna-
tional multicenter cohort study during the COVID-19
pandemic allowed international comparisons of short-
term outcomes between countries of different income
groups. In the presence of a low (1.4%) perioperative
COVID-19 rate, LLMIC was associated with almost 3
times higher odds of 30-day mortality compared to high
income. These ndings were not explained by patient
characteristics and postoperative pulmonary compli-
cations. The disparity in 30-day postoperative mortality
between different income countries should become a
focus of global neurosurgery and warrants further ex-
amination to identify any modiable factors that could
be addressed.
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

3
Poon et al.: Variation in postoperative neuro-oncology outcomes
Neuro-
Oncology
2-week windows from March to April 2020. We extracted
this for each participating hospital from the World Health
Organization, European Centre for Disease Prevention and
Control, or United States Centers for Disease Control and
Prevention statistics. We trichotomized patients into high
(≥58 cases per 100000 population) medium (10.3–58 cases
per 100000 population) and low (<10.3 cases per 100000
population) SARS-CoV-2 risk groups according to the pop-
ulation data at the time of surgery. COVID-19 free surgical
pathway referred to hospitals that utilized a system where
patients without SARS-CoV-2 infection underwent sur-
gery and perioperative care in hospital areas completely
separated from patients treated for COVID-19. When there
were changes to the intended management plan due to the
COVID-19 pandemic, the local study team recorded this
as a change from usual care. For example, these changes
may have occurred because of hospital bed shortages,
staff redeployment, or perceived high risk of SARS-CoV-2
infection risks. Other postoperative data included pulmo-
nary complications, which included pneumonia, acute res-
piratory distress syndrome, or unplanned postoperative
ventilation, and postoperative SARS-CoV-2 infection con-
firmed by a positive swab, positive thoracic CT imaging,
or a clinical diagnosis of symptomatic COVID-19 in patients
for whom these tests were unavailable. We also recorded
postoperative complications identified by the participating
hospitals.
Sample Size and Mitigation Against Bias
There was no sample size calculation for this exploratory
analysis of data generated from a rapid response research
collaborative. To account for potential bias that hospitals
more severely affected by Covid would participate, we col-
lected data on both community and postoperative COVID-
19 status. To minimize ascertainment bias, we requested
for additional validation of patient identification in hos-
pitals recruiting ≤5 patients. Testing and screening capacity
for COVID-19 was variable internationally during the study
period, which would introduce measurement bias of peri-
operative COVID-19 status. We used postoperative pulmo-
nary complication as a proxy variable to account for this
since pulmonary complication was less likely to be affected
by measurement bias in this 30-day study period.
Statistical Analyses
We used descriptive statistics to present characteristics of
patients in different income groups without univariable
analyses to avoid multiple testing. To account for the dif-
ferent operational characteristics of hospitals and the ex-
pected few number of deaths, we used Bayesian multilevel
logistic regression models with population stratification
by hospitals incorporated as random intercepts for our
multivariable analyses. Informative priors were based on
existing literature on the association between covariates
and 30-day postoperative mortality10–12 and experts in the
study group. Covariates in the multivariable model on
30-day mortality included the World Bank income groups,
age groups, sex, WHO performance status, ASA status,
urgency of surgery, and postoperative respiratory com-
plications. Sensitivity analysis using weakly informative
priors assessed the influence of informative priors on the
posterior distributions. Credible interval (CrI) represented
the 95% highest density interval of the posterior distribu-
tions, which can be interpreted as 95% confidence interval
but is philosophically distinct. WHO performance status
may have different prognostic value depending on the
context; a model including an interaction between income
groups and WHO performance status evaluated this poten-
tial effect modification. Interaction terms had weakly infor-
mative priors. We took a complete case analysis approach.
We accepted model as convergent if R-hat diagnostic was
<1.05. Other diagnostics checked for correct specification,
independence, and linearity. We performed all data hand-
ling and analyses in R (v4.1.0) using “tidyverse” (v1.3.1),
“gtsummary” (1.4.1), “brms” (v2.15.5), and “ROCR”
(v1.0-11) packages. We used “shinystan” (v.2.5.0) and
“loo” (v.4.2.1) for model parameters and convergence
diagnostics.
Results
Participating Hospitals
There were 1016 patients who underwent surgery for an in-
tracranial tumor in 55 participating hospitals from 26 coun-
tries. The 3-month patient recruitment periods across the
hospitals spanned between January 13, 2020 and August 9,
2020. Countries that contributed >50 patients were United
Kingdom (40.4%; N = 410), United States (9.7%; N = 99),
Saudi Arabia (9.5%; N=97), Serbia (7.7%; N=78), Morocco
(6.2%; N=63) and Italy (5.7%; N=58). There were 11 high-
income countries (HICs) contributing 765 (75.3%) patients,
7 upper-middle income countries (UMICs) contributing
142 (14.0%) patients, and 8 low and lower-middle income
countries (LLMICs) contributing 109 (10.7%) patients. The
median number of patients from each hospital during the
respective 3-month consecutive recruitment period was 7
(interquartile range [IQR] 2–34) patients.
Patient Characteristics
The proportions of patients in communities with low,
medium, and high SARS-CoV-2 risk were 26.8%, 35.9%,
and 36.6%, respectively (Table 1). Eleven (1.1%) patients
had confirmed or probable COVID-19 that had resolved
before the time of surgery, of which 9 occurred within 4
weeks preoperatively. Most (85.5%) underwent surgery
in hospitals without a COVID-19-free surgical pathway.
753 (74.1%) patients had preoperative Covid screening,
of which 551 (73.2%) had a swab test. There were 10 pa-
tients who tested positive using preoperative swab test
within 7 days. Most (83.3%) patients were <70 years of age
and 8.0% had a preexisting respiratory condition (Table
1). Gliomas were the most common tumor type (42.9 %)
followed by meningiomas (18.3%). Nine hundred eleven
(89.7%) patients had a tumor resection, and a gross total
resection was achieved in 521 (51.6%) patients.
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

4 Poon et al.: Variation in postoperative neuro-oncology outcomes
Table 1. Characteristics of 1016 Patients Who Underwent Surgery for an Intracranial Tumor
Income groups
Variables Overall
N = 1016
HIC
N = 765
UMIC
N = 142
LLMIC
N = 109
Community SARS-CoV-2 risk
Low 272 (26.8%) 129 (16.9%) 49 (34.5%) 94 (86.2%)
Medium 365 (35.9%) 291 (38.0%) 63 (44.4%) 11 (10.1%)
High 372 (36.6%) 343 (44.8%) 26 (18.3%) 3 (2.8%)
Unknown 7 (0.7%) 2 (0.3%) 4 (2.8%) 1 (0.9%)
Hospital type
COVID-19-free surgical pathway 147 (14.5%) 21 (2.7%) 100 (70.4%) 26 (23.9%)
Hospital with no defined pathway 869 (85.5%) 744 (97.3%) 42 (29.6%) 83 (76.1%)
Preoperative Covid screening 753 (74.1%) 551 (72.0%) 119 (83.8%) 83 (76.1%)
Age
<50 years 357 (35.1%) 255 (33.3%) 52 (36.6%) 50 (45.9%)
50–59 years 256 (25.2%) 180 (23.5%) 38 (26.8%) 38 (34.9%)
60–69 years 233 (22.9%) 184 (24.1%) 36 (25.4%) 13 (11.9%)
70–79 years 150 (14.8%) 129 (16.9%) 14 (9.9%) 7 (6.4%)
>80 years 20 (2.0%) 17 (2.2%) 2 (1.4%) 1 (0.9%)
Sex
Female 509 (50.1%) 387 (50.6%) 79 (55.6%) 43 (39.4%)
Male 507 (49.9%) 378 (49.4%) 63 (44.4%) 66 (60.6%)
BMI
Underweight 26 (2.6%) 20 (2.6%) 2 (1.4%) 4 (3.7%)
Normal 466 (45.9%) 297 (38.8%) 92 (64.8%) 77 (70.6%)
Overweight 317 (31.2%) 264 (34.5%) 31 (21.8%) 22 (20.2%)
Obese 204 (20.1%) 181 (23.7%) 17 (12.0%) 6 (5.5%)
Unknown 3 (0.3%) 3 (0.4%) 0 (0.0%) 0 (0.0%)
Preexisting respiratory condition 81 (8.0%) 72 (9.4%) 7 (4.9%) 2 (1.8%)
Current smoker 83 (8.2%) 67 (8.8%) 5 (3.5%) 11 (10.1%)
WHO performance status
0 418 (41.1%) 351 (45.9%) 38 (26.8%) 29 (26.6%)
1–2 514 (50.6%) 372 (48.6%) 80 (56.3%) 62 (56.9%)
3–4 78 (7.7%) 36 (4.7%) 24 (16.9%) 18 (16.5%)
Unknown 6 (0.6%) 6 (0.8%) 0 0
ASA grade
ASA grade 1–2 709 (69.8%) 492 (64.3%) 125 (88.0%) 92 (84.4%)
ASA grade 3–5 307 (30.2%) 273 (35.7%) 17 (12.0%) 17 (15.6%)
Urgency of surgery
Planned 406 (40.0%) 297 (38.8%) 65 (45.8%) 44 (40.4%)
Unplanned 610 (60.0%) 468 (61.2%) 77 (54.2%) 65 (59.6%)
Tumor location
Supratentorial 854 (84.1%) 632 (82.6%) 123 (86.6%) 99 (90.8%)
Infratentorial 162 (15.9%) 133 (17.4%) 19 (13.4%) 10 (9.2%)
Tumor type
Glioma 436 (42.9%) 319 (41.7%) 63 (44.4%) 54 (49.5%)
Meningioma 186 (18.3%) 143 (18.7%) 26 (18.3%) 17 (15.6%)
Primary CNS lymphoma 22 (2.2%) 13 (1.7%) 6 (4.2%) 3 (2.8%)
Vestibular schwannoma 29 (2.9%) 22 (2.9%) 5 (3.5%) 2 (1.8%)
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

5
Poon et al.: Variation in postoperative neuro-oncology outcomes
Neuro-
Oncology
More patients in LLMICs (87.0%) were in communities
with low SARS-CoV-2 risk than in UMICs (35.5%) and HICs
(16.9%). Of the 10 patients who were tested positive using
preoperative swab test, 1 and 9 patients were in UMIC
and HICs, respectively. Preexisting respiratory conditions
were more common in patients from HICs (9.4%) than in
those from UMICs (4.9%) or LLMICs (1.8%). Patients in
UMICs and LLMICs had worse WHO performance status
and better ASA grade, and those in LLMICs were younger
(Table 1). Tumor types of the patients were similar between
the 3 income groups. Gross total resection was achieved in
48.5% in HIC, 62.4% in UMIC, and 59.3% in LLMIC.
Pathology
There were 436 patients with a histopathologically confirmed
glioma and most (76.1%) had a grade 4 glioma. Overall,
41.7% patients had gross total resection of the glioma (Table
2). There were 1 oligoastrocytoma (not otherwise speci-
fied), 52 diffuse astrocytoma, 28 oligodendrogliomas, 18
anaplastic astrocytoma, and 336 glioblastoma. Of the 28 pa-
tients with oligodendroglioma, 17 (60.7%) had their gliomas
tested for both 1p/19q co-deletion and IDH mutation. In pa-
tients with a glioblastoma, 79.5% (n/N= 267/336) had IDH
mutation tested and 52.7% (n/N=177/336) had MGMT pro-
moter methylation status determined. Clinical and molec-
ular characteristics by income countries are presented in
Supplementary Tables 1–3.
Planned Treatment and Postoperative Outcomes
There was a change from the usual oncological care for
20.8% (N = 211) patients (Figure 1). The most common
change of care was a delay in surgical treatment (14.4%)
though 2.7% patients had their surgery expedited. There
were 26 (2.6%) patients who had a change to their planned
oncological treatment (Supplementary Table 4).
Within 30 days postoperatively, there were 44 (4.3%)
patients who had a respiratory complication and 14
(1.4%) patients who had a COVID-19 diagnosis. The
30-day postoperative mortality was 3.8%, which was
higher among patients in LLMICs (9.2%) than those in
UMICs (2.8%) and HICs (3.3%) (Figure 1). Of the 10 who
tested positive for Covid preoperatively, 2 patients
died within 30 days postoperatively. The 30-day mor-
tality of patients with and without a change to care was
3.0% and 4.0%, respectively. There was no difference in
postoperative complications between income groups
(Supplementary Table 5).
Income Groups and 30-day Mortality
Excluding 6 (0.6%) patients with incomplete data, we per-
formed our multivariable analyses on 30-day mortality using
data from 1010 patients. Patients in LLMICs had higher mor-
tality within 30 days after surgery compared to patients in
HICs (odds ratio [OR] 2.83, 95% CrI 1.37–5.74) (Figure 2).
There were no concerns with model convergence. A model
with the same covariates using weakly informative priors
centered on zero generated an OR of 2.68 (95% CrI 0.88–7.76),
indicating our informative prior regularized the variance of
the estimate without inflating the parameter estimate. There
was no evidence of higher mortality in UMICs (OR 1.24, 95%
CrI 0.32–4.62). We fitted a model with interaction between
income groups and WHO performance status. When com-
paring the leave-one-out cross validation of the models,
there was no evidence of better performance of the interac-
tion model (expected log pointwise predictive density differ-
ence [ELPD] was 0.5, standard error of ELPD difference was
–1.7).
Discussion
This study showed that in the first wave of the COVID-19
pandemic, about 1 in 5 neuro-oncology patients had a
change to their treatment plan from standard practice.
These changes were mostly related to the timing of sur-
gery rather than postsurgical oncological treatment. The
low SARS-CoV-2 risk in participating hospitals located in
LMICs allowed us to examine the characteristics and out-
comes of their patients with a relatively smaller impact
of the pandemic. This revealed that postoperative 30-day
mortality was higher in LLMICs compared to HICs.
Income groups
Variables Overall
N = 1016
HIC
N = 765
UMIC
N = 142
LLMIC
N = 109
Pituitary adenoma 74 (7.3%) 58 (7.6%) 8 (5.6%) 8 (7.3%)
Metastasis 147 (14.5%) 120 (15.7%) 17 (12.0%) 10 (9.2%)
Other 122 (12.0%) 90 (11.8%) 17 (12.0%) 15 (13.8%)
Extent of resection
Biopsy 105 (10.3%) 81 (10.6%) 8 (5.6%) 16 (14.7%)
Subtotal 316 (31.1%) 257 (33.6%) 36 (25.4%) 23 (21.1%)
Gross total 521 (51.3%) 369 (48.2%) 88 (62.0%) 64 (58.7%)
No postoperative imaging 68 (6.7%) 54 (7.1%) 9 (6.3%) 5 (4.6%)
Unknown 6 (0.6%) 4 (0.5%) 1 (0.7%) 1 (0.9%)
Table 1. Continued
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

6 Poon et al.: Variation in postoperative neuro-oncology outcomes
Impact of COVID-19 Pandemic
Two studies from the United Kingdom with overlapping re-
cruitment periods but both reflecting the first wave of the
pandemic reported 8.6–10.7% of patients who had their
management changed.5,6 However, there were variations
between neurosurgical centers from 0% to 28% that ap-
peared to correlate with the volume of neuro-oncology
patients.5 These studies also included all neuro-oncology
patients without restriction to those undergoing surgery.
Together with different hospital management strategies
adopted,13,14 these would explain the higher proportion of
patients with a change to their treatment plan in this study.
It is difficult to assess the impact of general surgical care
guidance15 and specific guidance for the management of
neuro-oncology patients during the pandemic.16–18 The de-
cision to alter usual care is a dynamic process that depends
on the volume of people with COVID-19 requiring hospi-
talization and the capacity of the hospitals for carrying out
medical and surgical oncological treatment in this context.
The effect of guidelines is likely to make decision-making
more similar between neuro-oncology services rather than
reducing the number of patients receiving non-routine
management.
Since the end of this study, there has been new evidence
about surgical care during the pandemic. The COVIDSurg-
Cancer study demonstrated that COVID-19-free surgical
pathways were associated with lower postoperative pul-
monary complications within 30 days.19 Planned delay of
surgery for 7 weeks in patients with preoperative SARS-
CoV-2 infection can mitigate the increased risk of short-term
pulmonary complication and perioperative mortality,20
though may not be possible or safe for patients with brain
tumors and raised intracranial pressure. Targeted use of
preoperative nasopharyngeal swab testing in areas with
high SARS-CoV-2 risk can be a strategy in lower resource
settings to implement measures to reduce the risk of post-
operative pulmonary complications.21 Although not re-
flected in our results, these findings can help reduce the
need to change routine care where possible.
Postoperative Mortality
Our 30-day mortality after surgery for a brain tumor in
UMICs and HICs was consistent to those in the published
literature.10–12,22,23 There is limited reporting of short-term
mortality in LLMIC. One study in Egypt—an LLMIC at the
time of writing—included 193 craniotomies for tumor re-
section over a 3-month period before the pandemic.24 They
reported a mean length of hospital stay of 9 days and an
in-hospital mortality of 10.5%. This supports the associ-
ation between LLMICs and higher 30-day mortality ob-
served in our study. It also suggests that this association
did not result from the COVID-19 pandemic since the com-
munity SARS-CoV-2 risk was low in LLMICs.
Surgical mortality is a leading cause of death globally
and 1 in 4 cancer patients receive a form of surgery as a
part of their cancer treatment. However, the quality of ev-
idence available on LLMICs is suboptimal due to selective
and poor reporting, thereby limiting comparisons.25 The
GlobalSurg initiative reported on the disparity in peri-
operative mortality in LLMICs compared to HIC.26 This
collaborative study included 15958 surgical patients with
primary colorectal, gastric, or breast cancer, of which 4131
were in LLMIC. In their multivariable analyses, 30-day
Table 2. Clinical and Molecular Characteristics of 436 Patients With a Glioma
Glioma
N (%)
Overall
N = 436
Grade 1
N = 20
Grade 2
N = 42
Grade 3
N = 42
Grade 4
N = 332
Extent of resection
Biopsy 72 (16.5%) 7 (35.0%) 8 (19.0%) 7 (16.7%) 50 (15.1%)
Subtotal 167 (38.3%) 4 (20.0%) 13 (31.0%) 6 (14.3%) 144 (43.4%)
Gross total 182 (41.7%) 7 (35.0%) 18 (42.9%) 27 (64.3%) 130 (39.2%)
Unknown 15 (3.4%) 2 (10.0%) 3 (7.1%) 2 (4.8%) 8 (2.4%)
1p/19q co-deletion
Intact 24 (5.5%) 0 (0.0%) 4 (9.5%) 3 (7.1%) 17 (5.1%)
Deleted 26 (6.0%) 1 (5.0%) 12 (28.6%) 9 (21.4%) 4 (1.2%)
Not tested/unknown 386 (88.5%) 19 (95.0%) 26 (61.9%) 30 (71.4%) 311 (93.7%)
IDH mutation
Wildtype 263 (60.3%) 3 (15.0%) 4 (9.5%) 12 (28.6%) 244 (73.5%)
Mutated 63 (14.4%) 2 (10.0%) 23 (54.8%) 15 (35.7%) 23 (6.9%)
Not tested/unknown 110 (25.2%) 15 (75.0%) 15 (35.7%) 15 (35.7%) 65 (19.6%)
MGMT promoter methylation
Unmethylated 103 (23.6%) 1 (5.0%) 4 (9.5%) 6 (14.3%) 92 (27.7%)
Methylated 94 (21.6%) 0 (0.0%) 3 (7.1%) 7 (16.7%) 84 (25.3%)
Not tested/unknown 239 (54.8%) 19 (95.0%) 35 (83.3%) 29 (69.0%) 156 (47.0%)
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

7
Poon et al.: Variation in postoperative neuro-oncology outcomes
Neuro-
Oncology
mortality was higher in LLMICs compared to HICs in pa-
tients following colorectal (OR 4.59, 95% CI 2.39–8.80) and
gastric (OR 3.72, 95% CI 1.70–8.16) cancer surgery. Their
findings suggested that a lower capacity to rescue major
complications associated with health system factors con-
tributed to the observed higher postoperative mortality.26
These factors such as access to imaging facilities and crit-
ical care facilities are shared with all surgical procedures.
Low & low-middle income
Any change to care
Upper-middle income
High income
28% (31/109)
15% (22/142)
21% (158/765)
Delayed surgery
Expedited surgery
Change in choice of operation
Surgery performed in another hospital
Respiratory complication
COVID-19 diagnosis
30-day mortality
18% (20/109)
13% (18/142)
14% (108/765)
7% (8/109)
2% (3/142)
2% (16/765)
3% (3/109)
1% (1/142)
1% (7/765)
0% (0/109)
0% (0/142)
3% (20/765)
7% (8/109)
1% (2/142)
4% (34/765)
1% (1/109)
0% (0/142)
2% (13/765)
9% (10/109)
3% (4/142)
3% (25/765)
0510 15
Proportion of patients
(n/N)
20 25 30
Low & low-middle income
Upper-middle income
High income
Low & low-middle income
Upper-middle income
High income
Low & low-middle income
Upper-middle income
High income
Low & low-middle income
Upper-middle income
High income
Low & low-middle income
Upper-middle income
High income
Low & low-middle income
Upper-middle income
High income
Low & low-middle income
Upper-middle income
High income
Figure 1. Change in treatment and postoperative outcomes by income groups. Each panel represents the proportion of patients in which the
event occurred. Respiratory complications included pneumonia, acute respiratory distress syndrome, and unplanned mechanical ventilation
postoperatively. The numbers to the right of each bar are the percentage and number of events over the number of patients in each income
groups.
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

8 Poon et al.: Variation in postoperative neuro-oncology outcomes
It is plausible that these factors could explain the higher
30-day mortality of patients with a brain tumor in LLMICs.
Strengths and Limitations
This study had low levels of missing data and included
common prognostic variables associated with short-term
outcomes. Our findings provided data on short-term
neuro-oncology surgical outcomes after operations for in-
tracranial tumors across different countries. We were able
to describe the use of molecular markers in glioma diag-
nosis in different settings. Importantly, our data added to
the limited literature on early postoperative outcomes in
LLMIC.
We were unable to include preoperative positive swab
results into our model because of data sparsity, with no pa-
tients in LLMICs having a positive test. Although this may
lead to an underestimation of the effect size for mortality
associated with LLMIC, this does not change our narrative
of the disparity in postoperative outcome. This study did
not collect information on the cause of death. This would
have been useful to identify preventable deaths following
surgery. We accounted for variations in hospital path-
ways using a hierarchical model, but there was likely to
be residual confounding from the healthcare system and
infrastructure that can affect postoperative outcomes.
However, these effects are likely to be small since our es-
timate is similar to larger studies examining general sur-
gical outcomes. A quarter of the patients did not have
a preoperative COVID-19 screening and postoperative
SARS-CoV-2 infection may be variably reported due to
testing capacity. But this was unlikely to affect our results
because we used pulmonary complications as a variable in
our model, which would account for the higher risk of pul-
monary complications associated with preoperative SARS-
CoV-2 infection. It was not possible to determine whether
the delay in surgery affected the short-term postoperative
mortality. Because patients with delayed surgery can have
deterioration in WHO performance status, controlling for
WHO performance status in our model would, at least par-
tially, account for the delay in surgery. Lastly, we were un-
able to compare our results to those from an equivalent
precovid dataset to determine and extrapolate the effect
of the COVID-19 pandemic more confidently. Details about
case-volume, case-mix, surgical preparedness27, and back-
ground surgical outcomes can help to interpret differences
in postoperative mortality.
Conclusions
The impact of the COVID-19 pandemic on neurosurgical
services for patients with intracranial tumors mainly af-
fected surgical care and there was a low (0.3%) propor-
tion of patients having SARS-CoV-2 infection within 30
days after surgery in the participating centers during
the pandemic’s first wave. Postoperative mortality was
higher in LLMICs than in UMICs and HIC, which was not
explained by patient characteristics and postoperative
pulmonary complications. As the pandemic evolves and
new evidence becomes available for the management of
surgical patients, neuro-oncology centers can adopt the
safest surgical pathways for their patients and audit their
performances against the findings presented in this study.
High income
AB
Upper middle income
Low and lower middle income
Age <50 years
Age 50–59 years
Age 60–69 years
Age 70–79 years
Age 80+ years
Female
Male
ECOG 0
ECOG 1–2
ECOG 3–4
ASA 1–2
ASA 3–5
Planned surgery
Unplanned surgery
No respiratory complication
Respiratory complication
124
Odds ratio
816
Reference
1.49 (0.68–3.29)
2.83 (1.37–5.74)
Reference
1.56 (0.84–2.89)
1.72 (0.90–3.26)
1.69 (0.84–3.35)
2.82 (1.20–6.54)
Reference
1.32 (0.71–2.51)
Reference
1.48 (0.82–2.69)
1.93 (0.94–3.97)
Reference
1.74 (1.02–2.93)
Reference
1.75 (1.01–3.07)
Reference
4.00 (1.95–7.99)
High income
Upper middle income
Low and lower middle income
Age <50 years
Age 50–59 years
Age 60–69 years
Age 70–79 years
Age 80+ years
Female
Male
ECOG 0
ECOG 1–2
ECOG 3–4
ASA 1–2
ASA 3–5
Planned surgery
Unplanned surgery
No respiratory complication
Respiratory complication
124
Odds ratio
816
Reference
1.24 (0.32–4.62)
2.68 (0.88–7.76)
Reference
1.45 (0.61–3.51)
1.43 (0.55–3.67)
1.20 (0.39–3.52)
2.67 (0.42–13.41)
Reference
1.38 (0.70–2.76)
Reference
1.11 (0.51–2.48)
1.27 (0.36–4.22)
Reference
1.59 (0.72–3.49)
Reference
1.20 (0.58–2.57)
Reference
9.01 (3.51–22.55)
Figure 2. Bayesian multilevel logistic regression models on 30-day postoperative mortality in 1010 patients. (A) Forest plot of odds ratios from
the Bayesian multilevel logistic regression on 30-day postoperative mortality using informative priors. (B) Forest plot showing odds ratios from
sensitivity analysis using weakly informative priors to assess the inuence of informative priors on the posterior distributions of the odds ratios.
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

9
Poon et al.: Variation in postoperative neuro-oncology outcomes
Neuro-
Oncology
Disparity in 30-day postoperative mortality between dif-
ferent income countries should become a focus of global
neurosurgery and warrants further examination to identify
any modifiable factors that could be addressed.
Supplementary material
Supplementary material is available online at Neuro-
Oncology (http://neuro-oncology.oxfordjournals.org/).
Keywords
collaborative research | global neurosurgery |
neuro-oncology
Acknowledgments
This research was part-funded by the National Institute for
Health Research (NIHR; NIHR 16.136.79) using UK aid from
the UK Government to support global health research. RS re-
ceives funding from the Economic and Social Research Council.
JCG and AAB are funded by personal awards from the NIHR
Academy. MTCP was funded by Cancer Research UK Brain
Tumour Centre of Excellence Award (C157/A27589). HJM was
funded by the Wellcome/ EPSRC Centre for Interventional and
Surgical Sciences and NIHR Biomedical Research Centre at
University College London. The views expressed in this publica-
tion are those of the authors and not necessarily those of the
NIHR or the UK Government.
Collaborators (listed by country in alphabetical order;
asterisk denotes local principal investigator)Austria: J.
Burtscher, F. Trivik-Barrientos* (Landesklinikum Wiener
Neustadt, Wiener Neustadt); M. Bauer, C. Freyschlag (Medical
University Of Innsbruck, Innsbruck). Brazil: M. Lemos Vieira
da Cunha* (Supera Oncologia - Hospital Regional Do Oeste,
Chapeco). Canada: G. Groot*, A. Persad, H. Pham, M. Wood
(Saskatoon City Hospital/Royal University Hospital/St. Paul’s
Hospital, Saskatoon Sk). Denmark: P. Christensen*, M.
Haldrup, L.H. Iversen, H.Ø. Kristensen, M. Mekhael, N. Mikic
(Aarhus University Hospital, Aarhus). Dominican Republic:
A. Crespo*, P. Díaz, N. Tactuk (Cedimat - Centro De Diagnóstico,
Medicina Avanzada, Laboratorio Y Telemedicina, Santo
Domingo). Egypt: A. Abdelsamed, A.Y. Azzam*, H. Salem*, A.
Seleim (Al Azhar University Hospitals, Cairo); S. Abd-elsalam, H.
Badr, M. Elbahnasawy*, M. Essa, S. Gamal Badr, A. Ghoneim*,
O. Hamad, M. Hamada, A. Hawila, M.S. Morsy, S. Sarsik (Tanta
University Hospital, Tanta). France: Q. Ballouhey *, H. Salle
(Chu Limoges, Limoges). Guatemala: A. Barrios Duarte*, I.
Lopez Muralles, M. Lowey, A.L. Portilla, G. Recinos (Hospital
General De Enfermedades, Guatemala City). India: R. Arora*, R.
Kottayasamy Seenivasagam*, S. Sadhasivam (All India Institute
Of Medical Sciences, Rishikesh); N. Babu, Y. Kheni, V. Kommu,
S. Rao* (Basavatarakam Indo American Cancer Hospital &
Research Institute, Hyderabad); A. Moiyadi, D. Pandey, C.S.
Pramesh*, P. Shetty, V. Singh (Tata Memorial Hospital, Mumbai).
Indonesia: A.A. Islam*, G. Kembuan, H. Pajan (Rsud Wahidin
Sudirohusodo, Makassar). Iran: H. Safari (Golestan Hospital,
Ahvaz). Italy: F. Bàmbina, G. D’Andrea, P. Familiari*, V. Picotti
(Fabrizio Spaziani, Frosinone 03100); P. Bruzzaniti, V. Chiarella, A.
Di bartolomeo, A. Frati, M. Giugliano, P. Lapolla*, M. Salvati, A.
Santoro, A.K. Scafa (Policlinico Umberto I Sapienza University
of Rome, Rome); F. Gagliardi, M. Medone, P. Mortini*, M. Piloni
(San Raffaele Scientic Institute, Milan, Milan); A. Belvedere,
M. Droghetti, F. Frio*, J. Neri, A.P. Pezzuto, G. Poggioli, M.
Rottoli*, I.S. Russo (Sant’orsola Hospital, Alma Mater Studiorum
University Of Bologna, Italy, Bologna); F. Aquila, C. Gambacciani,
L. Lippa, F. Pieri, O.S. Santonocito* (Spedali Riuniti Di Livorno,
Livorno). Jordan: M. Al Abdallah*, F. Ayasra, Y. Ayasra, A.
Qasem (Al-Basheer Hospital, Amman); F.J. Abu Za’nouneh, A.A.
Al_shraideh, T. Fahmawee, A. Ibrahim (King Abdullah University
Hospital, Ar Ramtha); M. K. Abou Chaar, H. Al-Najjar, M. Elayyan
(King Hussein Cancer Center, Amman). Libya: M. Abusannoga,
A. Alawami, M. Alawami*, M. Albashri, A. Malek (Medical Care
Clinic, Tripoli); E. Abdulwahed*, M. Biala, R. Ghamgh (Tripoli
Central Hospital, Tripoli). Morocco: Y. Arkha, H. Bechri, A. El
Ouahabi, M.Y. Oudrhiri* (Centre Hospitalier Universitaire Ibn
Sina Rabat, Rabat); A. El Azhari, S.M. Louraoui*, M. Rghioui
(Cheikh Khalifa International University Hospital, Casablanca
City); M. Bougrine, F. Derkaoui Hassani*, N. El Abbadi (Cheikh
Zaid International University Hospital, Rabat). Nigeria: A.
Akinmade*, S. Fayose (Afe Babalola University Multi-System
Hospital, Ido Ekiti); A. Okunlola* (Federal Teaching Hospital,
Ido Ekiti, Ido Ekiti); Y. Dawang, J. Obande, S. Olori* (University
of Abuja Teaching Hospital, Gwagwalada); L. Abdur-Rahman*,
N. Adeleke, A. Adeyeye* (University of Ilorin Teaching Hospital,
Ilorin). Pakistan: S. Javed*, E. Yaqoob* (Holy Family Hospital,
Rawalpindi). Palestine: I. Al-Slaibi, H. I. A. Alzeerelhouseini, F.
Jobran* (Al-Ahli Hospital, Hebron, West Bank). Saudi Arabia:
M. Alshahrani*, F. Alsharif (Aseer Central Hospital, Abha); M.
A. Azab* (King Abdullah Medical City Makkah, Makkah); F.
Al Otaibi, H. AlDahash, N. Alhazzaa, A. Alhefdhi*, T. AlSumai,
F. Farrash, P. Spangenberg (King Faisal Specialist Hospital,
Riyadh); A. Ajlan, A. Al-Habib, A. Alatar, A. Bin Nasser*, S.
Elwatidy, T. Nouh* (King Saud University, Riyadh); F. Abdulfattah,
F. Alanazi, F. Albaqami, K. Alsowaina (Prince Sultan Military
Medical City, Riyadh). Serbia: V. Bascarevic, I. Bogdanovic, D.
Grujičić *, R. Ilic*, M. Milićević, F. Milisavljević, A. Miljković, A.
Paunovic, V. Šćepanović, A. Stanimirovic, M. Todorovic (Clinic
for Neurosurgery, Clinical Center of Serbia, Belgrade). Spain:
A.M. Castaño-Leon*, J. Delgado Fernandez, C. Eiriz Fernandez,
O. Esteban Sinovas, D. Garcia Perez, P. Gomez, L. Jimenez-
Roldan, A. Lagares, L. Moreno-Gomez, I. Paredes, A. Pérez
Núñez (12 De Octubre University Hospital, Madrid); I. Aldecoa
Ansorregui, A. Di Somma*, J. Enseñat Nora*, N. Fabregas, A.
Ferrés, J.J. Gonzalez Sanchez*, I. Gracia, J.A. Hoyos Castro, C.
Langdon, L. Oleaga, L. Pedrosa, J. Poblete Carrizo, L.A. Reyes
Figueroa, P. Roldan Ramos, J. Rumia-Arboix, A.I. Tercero-Uribe,
T.E. Topczewski, J. Torales, R. Torné, R. Valero (Hospital Clinic
Barcelona, Barcelona). Syrian Arab Republic: M. Mahfoud
(Tishreen University Hospital, Latakia). United Kingdom: M.
Bekheit* (Aberdeen Royal Inrmary, Aberdeen); J. Ashcroft,
P. Coughlin, R.J Davies*, P. Hutchinson*, D.Z. Khan, A. Kolias,
R. Mannion, M. Mohan, S. Price, T. Santarius, A. Singh, S.
Yordanov (Addenbrooke’s Hospital, Cambridge); M. Ganau*,
D. Jeyaretna, R. Piper*, S. Sravanam (John Radcliffe Hospital,
Oxford); N. McSorley, A. Solth (Ninewells Hospital, Dundee); Y.
Chowdhury*, K. Karia, G. Solomou, W.C. Soon, A. Stevens, C.
Topham, I. Ughratdar (Queen Elizabeth Hospital Birmingham,
Birmingham); L. Alakandy, P. Bhattathiri, J. Brown, M. Canty,
A. Grivas, S. Hassan, S. Lammy*, P. Littlechild, C. Maseland,
C. Mathieson, R. O’Kane, E. St. George, N. Suttner, W. Taylor
(Queen Elizabeth University Hospital, Glasgow); Y. Al-Tamimi,
A. Bacon, M. Crank, O. Rominiyi*, S. Sinha (Royal Hallamshire
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

10 Poon et al.: Variation in postoperative neuro-oncology outcomes
Hospital, Shefeld); P.M. Brennan*, R. Pasricha (Royal Inrmary
Of Edinburgh, Edinburgh); A. Anzak, I. Leal Silva, C. Sohrabi, B.
Thakur * (Royal London Hospital, London); P. Patkar, I. Phang*
(Royal Preston Hospital, Preston); F. Colombo, D. Fountain, M.T.
Hasan, K. Karabatsou*, R. Laurente, O. Pathmanaban* (Salford
Royal Hospital, Salford); D. Choi, R. Hutchison, A. Jain, V. Luoma,
H. Marcus*, R. May, A. Menon, B. Pramodana, L. Webber (The
National Hospital for Neurology and Neurosurgery, London);
T. Elmoslemany, M. Jenkinson*, C. P. Millward, R. Zakaria (The
Walton Centre NHS Foundation Trust, Liverpool). United
States of America: B. Bigelow, E. Etchill*, A. Gabre-Kidan*, H.
Jenny, M. Ladd, C. Long, H. Malapati, A. Margalit, S. Rapaport,
J. Rose, L. Tsai, D. Vervoort, P. Yesantharao (Johns Hopkins
Hospital, Baltimore, MD); G. Arzumanov, N. Glass*, K. Zhao
(The University Hospital, Newark, NJ); S. Aoun, V.S. Ban*, H.H.
Batjer, J. Caruso (University Of Texas Southwestern, Dallas);
N.M. Ruzgar, M. Sion, S. Ullrich (Yale New Haven Hospital, New
Haven, CT).
Funding
National Institute for Health Research Global Health Research
Unit, Association of Coloproctology of Great Britain and
Ireland, Bowel and Cancer Research, Bowel Disease Research
Foundation, Association of Upper Gastrointestinal Surgeons,
British Association of Surgical Oncology, British Gynaecological
Cancer Society, European Society of Coloproctology, Medtronic,
Sarcoma UK, The Urology Foundation, Vascular Society for
Great Britain and Ireland, and Yorkshire Cancer Research.
Conflict of Interest
All authors declare no conict of interest.
Authorship
The writing group contributed to the writing, data interpreta-
tion, and critical revision of the manuscript. The collaborators
contributed to data collection and study governance across in-
cluded sites.
Affiliations
Usher Institute, University of Edinburgh, Edinburgh, UK (MTCP);
Department of Clinical Neurosciences, Royal Inrmary of
Edinburgh, Edinburgh, UK (M.T.C.P., A.K.D.); Department of
Neurosurgery, John Radcliffe Hospital, Oxford, UK (R.J.P.);
Department of Surgery, Division of Neurosurgery, University of
Cape Town, Cape Town, South Africa (N.T.); Manchester Centre
for Clinical Neurosciences, Salford Royal NHS Foundation
Trust, Salford, UK (D.M.F.); National Hospital for Neurology and
Neurosurgery, London, UK and UCL Queen Square Institute of
Neurology, London, UK (H.J.M.); Department of Neurosurgery,
Grande Ospedale Metropolitano Niguarda Milan, Italy (L.L.);
Department of Neurosurgery, Humanitas University, Milano,
Italy (F.S.); Neurosurgery Division, Faculty of Health Sciences,
University of Bamenda, Bambili, Cameroon (I.N.E.); Department
of Neurosurgery, Medical University of Innsbruck, Anichstr. 35,
6020 Innsbruck, Austria (C.F.F.); Instituto do Cancer do Estado de
Sao Paulo, Hospital das Clinicas da Faculdade de Medicina da
Universidade de Sao Paulo, Sao Paulo, Brazil (I.S.N.); George
Washington University School of Medicine and Health Sciences,
Washington, DC, United States (G.R.); Department of Clinical
Neurosciences, Geneva University Medical Center, Geneva,
Switzerland (K.S.); Department of Neurosurgery, Massachusetts
General Hospital, Boston, Massachusetts, United States (F.C.R.);
Academic Division of Neurosurgery, Addenbrooke's Hospital,
Cambridge, UK (P.J.H.); Department of Clinical Neurosciences,
Division of Neurosurgery, Cambridge, UK (S.J.P., A.G.K.); Division
of Neurosurgery, Department of Neurosciences, Philippine
General Hospital, University of the Philippines Manila, Manila,
Philippines (R.E.P.); NIHR Global Health Research Unit on Global
Surgery, University of Birmingham, Birmingham, UK (J.C.G.);
University of Birmingham, Birmingham, UK (A.B.); Department
of Neurosurgery, Walton Centre & University of Liverpool,
Liverpool, UK (M.D.J.)
References
1. COVIDSurg C. Elective surgery cancellations due to the COVID-19 pan-
demic: global predictive modelling to inform surgical recovery plans:
elective surgery during the SARS-CoV-2 pandemic. Br J Surg. Published
online June 13, 2020.
2. Nepogodiev D, Bhangu A, Glasbey JC, et al. Mortality and pulmo-
nary complications in patients undergoing surgery with periopera-
tive SARS-CoV-2 infection: an international cohort study. Lancet.
2020;396(10243):27–38.
3. Kuderer NM, Choueiri TK, Shah DP, et al. Clinical impact of COVID-
19 on patients with cancer (CCC19): a cohort study. Lancet.
2020;395(10241):1907–1918.
4. Lai AG, Pasea L, Banerjee A, et al. Estimated impact of the COVID-19
pandemic on cancer services and excess 1-year mortality in people with
cancer and multimorbidity: near real-time data on cancer care, cancer
deaths and a population-based cohort study. BMJ Open. 2020;10(11):e04
3828e043828.
5. Price SJ, Joannides A, Plaha P, et al. Impact of COVID-19 pan-
demic on surgical neuro-oncology multi-disciplinary team deci-
sion making: a national survey (COVID-CNSMDT Study). BMJ Open.
2020;10(8):e040898.
6. Fountain DM, Piper RJ, Poon MTC, et al. CovidNeuroOnc: a UK
multicenter, prospective cohort study of the impact of the COVID-
19 pandemic on the neuro-oncology service. Neuro-Oncol Adv.
2021;3(1):vdab014.
7. Amoo M, Horan J, Gilmartin B, et al. The provision of neuro-oncology
and glioma neurosurgery during the SARS-CoV-2 pandemic: a single
national tertiary centre experience. Ir J Med Sci. 2021;190(3):905–911.
8. Dewan MC, Rattani A, Fieggen G, et al. Global neurosurgery: the
current capacity and decit in the provision of essential neurosur-
gical care. Executive Summary of the Global Neurosurgery Initiative
at the Program in Global Surgery and Social Change. J Neurosurg.
2019;130(4):1055–1064.
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023

11
Poon et al.: Variation in postoperative neuro-oncology outcomes
Neuro-
Oncology
9. Glasbey J, Ademuyiwa A, Adisa A, et al. Effect of COVID-19 pan-
demic lockdowns on planned cancer surgery for 15 tumour types in 61
countries: an international, prospective, cohort study. Lancet Oncol.
2021;22(11):1507–1517.
10. Lassen B, Helseth E, Rønning P, et al. Surgical mortality at 30 days and
complications leading to recraniotomy in 2630 consecutive craniotomies
for intracranial tumors. Neurosurgery. 2011;68(5):1259–1269.
11. Williams M, Treasure P, Greenberg D, Brodbelt A, Collins P.; on behalf
of the (UK) National Cancer Information Network Brain Tumour Group.
Surgeon volume and 30 day mortality for brain tumours in England. Br J
Cancer. 2016;115(11):1379–1382.
12. Senders JT, Muskens IS, Cote DJ, et al. Thirty-day outcomes after cra-
niotomy for primary malignant brain tumors: a national surgical quality
improvement program analysis. Neurosurgery. 2018;83(6):1249–1259.
13. Hameed NUF, Ma Y, Zhen Z, et al. Impact of a pandemic on surgical
neuro-oncology—maintaining functionality in the early phase of crisis.
BMC Surg. 2021;21(1):40.
14. Mrugala MM, Ostrom QT, Pressley SM, et al. The state of neuro-
oncology during the COVID-19 pandemic: a worldwide assessment.
Neuro-Oncol Adv. 2021;3(1):vdab035.
15. COVIDSurg C, Bhangu A, Lawani I, et al. Global guidance for surgical
care during the COVID-19 pandemic. Br J Surg. 2020;107(9):1097–1103.
16. Bernhardt D, Wick W, Weiss SE, et al. Neuro-oncology management
during the COVID-19 pandemic with a focus on WHO grades III and IV
gliomas. Neuro-Oncology. 2020;22(7):928–935.
17. Ramakrishna R, Zadeh G, Sheehan JP, Aghi MK. Inpatient and outpatient
case prioritization for patients with neuro-oncologic disease amid the
COVID-19 pandemic: general guidance for neuro-oncology practitioners
from the AANS/CNS Tumor Section and Society for Neuro-Oncology. J
Neurooncol. 2020;147(3):525–529.
18. Society for British Neurological Surgeons, British Neuro-Oncology
Society. Adult Neuro-Oncology Service Provision During COVID. https://
www.bnos.org.uk/clinical-practice/treatment-pathways-and-guidance/.
Accessed September 8, 2022.
19. Glasbey JC, Nepogodiev D, Simoes JFF, et al. Elective cancer surgery
in COVID-19–free surgical pathways during the SARS-CoV-2 pandemic:
an international, multicenter, comparative cohort study. J Clin Oncol.
2021;39(1):66–78.
20. COVIDSurg C, GlobalSurg C, Nepogodiev D, et al. Timing of surgery fol-
lowing SARS-CoV-2 infection: an international prospective cohort study.
Anaesthesia. 2021;76(6):748–758.
21. COVIDSurg C, Glasbey JC, Omar O, et al. Preoperative nasopharyngeal
swab testing and postoperative pulmonary complications in patients
undergoing elective surgery during the SARS-CoV-2 pandemic. Br J
Surg. 2021;108(1):88–96.
22. De Witt Hamer PC, Ho VKY, Zwinderman AH, et al. Between-hospital
variation in mortality and survival after glioblastoma surgery in the Dutch
Quality Registry for Neuro Surgery. J Neurooncol. 2019;144(2):313–323.
23. Perla KMR, Pertsch NJ, Leary OP, et al. Outcomes of infratentorial cranial
surgery for tumor resection in older patients: an analysis of the National
Surgical Quality Improvement Program. Surg Neurol Int. 2021;12:144.
24. Helal AE, Abouzahra H, Fayed AA, Rayan T, Abbassy M. Socioeconomic
restraints and brain tumor surgery in low-income countries. Neurosurg
Focus. 2018;45(4):E11.
25. Ng-Kamstra JS, Arya S, Greenberg SLM, et al. Perioperative mortality
rates in low-income and middle-income countries: a systematic review
and meta-analysis. BMJ Glob Health. 2018;3(3):e000810.
26. Knight SR, Shaw CA, Pius R, et al. Global variation in postoperative mor-
tality and complications after cancer surgery: a multicentre, prospective
cohort study in 82 countries. Lancet. 2021;397(10272):387–397.
27. Glasbey JC, Abbott TE, Ademuyiwa A, et al. Elective surgery system
strengthening: development, measurement, and validation of the sur-
gical preparedness index across 1632 hospitals in 119 countries. Lancet.
2022;400(10363):1607–1617.
Downloaded from https://academic.oup.com/neuro-oncology/advance-article/doi/10.1093/neuonc/noad019/7117593 by guest on 15 April 2023