AJTCCM VOL. 29 NO. 1 2023 18
RESEARCH
Background. Mortality rates in patients with haematological malignancies who required intensive care unit (ICU) admission have in the
past been high. More recently, however, improved outcomes for critically ill haematological patients have been reported.
Objective. To determine outcomes, average length of ICU stay, and factors associated with mortality in patients with haematological
malignancies and neutropenic fever in the multidisciplinary ICU (MICU) at Universitas Academic Hospital (UAH), Bloemfontein, Free
State Province, South Africa.
Methods. We conducted a retrospective review of medical and laboratory records of all patients admitted to the UAH MICU with
haematological malignancies and febrile neutropenia between 2010 and 2019.
Results. A total of 182 patients with haematological malignancies were admitted to the MICU between 1 January 2010 and 31 December
2019, of whom 51 (28.0%) fullled the inclusion criteria for the study. e median age was 33 years, and 29 patients (56.9%) were female.
Most patients had either acute myeloid leukaemia (n=22; 43.1%) or acute lymphocytic leukaemia (n=16; 31.4%), while B-cell lymphoma
(n=12; 23.5%) and multiple myeloma (n=1; 2%) were less frequent. e median length of stay in the ICU was 3 days. ICU mortality
was 76.5% and hospital mortality 82.4%. Factors associated with mortality included septic shock, vasoactive agent use and mechanical
ventilation.
Conclusion. Patients with haematological malignancies and febrile neutropenia in the UAH MICU have high ICU and hospital mortality
rates. More needs to be done with regard to timeous management of patients with haematological malignancies and septic shock in our
setting to improve survival.
Keywords. Haematological malignancies, neutropenic fever, sepsis, intensive care unit.
Afr J Thoracic Crit Care Med 2023;29(1):e263. https://doi.org/10.7196/AJTCCM.2023.v29i1.263
Study synopsis
is is the rst study to report on ICU mortality of adult patients with haematological malignancies and neutropenic sepsis in a tertiary
hospital ICU in the Free State. ese patients had a high mortality rate.
What the study adds. Our study shows that septic shock, vasoactive agent use and mechanical ventilation were associated with increased
ICU mortality.
Implications of the ndings. Strict adherence to infection prevention and control measures in haematology wards is required. Early
recognition and treatment of sepsis before it progresses to septic shock is important. ICUs must be designed so that isolation cubicles are
readily available to prevent cross-infection of patients.
The World Health Organization estimated that there were ~1.28
million cases of haematological malignancy globally in 2020, with
710 000 deaths.[1] During the same year, an estimated 109000 cases
of haematological malignancies were diagnosed in South Africa (SA),
and ~70500 of these patients died.[1] Mortality rates in patients with
haematological malignancies who required intensive care unit (ICU)
admission have in the past been high.[2] More recently, however,
improved outcomes for critically ill haematological patients have been
published,[3] with hospital survival rates of up to 60.7% reported for a
European cohort of patients.[4] It is not surprising that an increasing
number of patients are referred and accepted for ICU admission.[5,6]
In resource-limited settings such as SA, conditions are oen very
dierent from developed-world settings. Decisions on ICU admission
invariably use triage and prioritisation models based on the improved
incremental benet that is highly inuenced by prognostication and
decision-making support systems.[7] Multiple-organ failure, the
requirement for invasive mechanical ventilation or vasopressors,
and Acute Physiology and Chronic Health Evaluation (APACHE II)
Outcomes of patients with haematological malignancies and febrile
neutropenia at the Universitas Academic Hospital multidisciplinary
intensive care unit, Free State Province, South Africa
C D S Martins,1 MB ChB ; S D Maasdorp,1,2 MB ChB, MMed (Int Med), FCP (SA), Cert Pulmonology (SA) Phys
1 Division of Critical Care, Department of Surgery, Faculty of Health Sciences, University of the Free State, Bloemfontein, South Africa
2 Division of Pulmonology, Department of Internal Medicine, Faculty of Health Sciences, University of the Free State, Bloemfontein, South Africa
Corresponding author: C D S Martins (martinsc@ufs.ac.za)
AJTCCM VOL. 29 NO. 1 2023 19
RESEARCH
and Sequential Organ Failure Assessment (SOFA) scores have been
found to be predictors of poor outcome in critically ill haematological
patients.[8-11] However, it remains important that ICU admission
policies reect local data, that patients who are too ill to benet are
not inappropriately admitted for prolonged organ support, and that
patients who may benet are not inadvertently denied intensive care
management.
At Universitas Academic Hospital (UAH), a tertiary hospital in
Free State Province, SA, patients with underlying haematological
malignancies undergoing chemotherapy are oen referred for ICU
admission aer developing febrile neutropenia with septic shock and/
or organ failure. Given the reported globally improved survival rates in
patients with haematological malignancies, we aimed to determine the
outcomes of patients with haematological malignancies with febrile
neutropenia admitted to the UAH multidisciplinary ICU (MICU)
between January 2010 and December 2019. e primary objective was
to determine the mortality rate of the patients. Secondary objectives
were to determine the median length of stay in the MICU, and identify
prognostic variables that were associated with poor outcome.
Methods
Study design
is was a retrospective descriptive study. Ethics approval to conduct
the study was obtained from the Health Sciences Research Ethics
Committee of the University of the Free State (UFS) (ref. no. UFS-
HSD2021/0186/2505).
Setting
UAH is a 636-bed hospital in Bloemfontein that serves as the
referral hospital for the Free State. All patients with haematological
malignancies in the Free State, Northern Cape Province and Lesotho
are referred to the haematology division at UAH for management.
Patients who develop neutropenic fever while undergoing
chemotherapy for their underlying haematological malignancy are
frequently referred to the UAH MICU for further management.
Study participants
Patients were included in the study if they were admitted to the UAH
MICU with a diagnosis of neutropenic fever, had a concomitant
diagnosis of an underlying haematological malignancy, and were
admitted between 1 January 2010 and 31 December 2019. Patients
with a haematological malignancy but for whom the primary reason
for referral to the UAH MICU was not neutropenic fever were
excluded.
Data collection
A data form was designed to capture information on age, sex, type
of haematological malignancy, reason for ICU admission (ventilator
support required, management of circulatory shock), severity of
illness scores at the time of admission to the ICU (APACHE II and
SOFA), HIV status (including absolute CD4 count and HIV viral
load for HIV-positive patients, if available), ratio of arterial oxygen
partial pressure to fractional inspired oxygen (PaO2/FiO2), white
cell count, neutrophil count, platelet count, Glasgow Coma Scale
(GCS) score (out of 15), total bilirubin level, serum creatinine level,
lactate level, systolic blood pressure, any requirement for vasoactive
agents at or prior to ICU admission, blood culture results (taken
within 72hours before or during the ICU stay), vasoactive agents
(adrenaline, phenylephrine, noradrenaline, dobutamine) used at any
time during the ICU stay, length of ICU stay (calculated from date
of admission to date of discharge from the ICU), ICU survival and
hospital survival.
Medical les of patients were reviewed. A pilot study with three
patients was performed to test all aspects of the data form. No
signicant changes to the data form were required, and the results of
the pilot study were included in the main study.
Bias
As this was a retrospective study of patient les, the medical records
may have been incomplete, and information on all the data required
to complete the data forms may not have been available. All patients in
whom the primary outcome could be determined were included in the
study. Patients were excluded from analysis pertaining to secondary
outcomes if the information required to determine secondary
outcomes was not available in the clinical les.
Bias could have been introduced in determining primary and
secondary outcomes in this study by the lack of ICU bed availability
in public sector hospitals, resulting in inability to accommodate all
patients with haematological malignancies who developed neutropenic
fever and were referred to the ICU. e number of patients who could
not be accommodated due to limited ICU bed availability should,
however, be small. We addressed this potential bias by increasing the
duration of the study period to 10 years to include as many patients
as possible in the study.
Another factor that could have resulted in bias in terms of the
primary outcome is that treatment may have been withdrawn
from patients when further treatment was deemed futile. is is,
however, a universally accepted management strategy in patients
with refractory organ failure in whom all other curative treatment
options have been exhausted and whose likelihood of survival is
deemed extremely poor.
Data analysis
Continuous variables were summarised by medians, minimum,
maximum or percentiles. Categorical variables were summarised
by frequencies and percentages. Dierences between groups were
evaluated using appropriate statistical tests (χ2 test or Fisher’s exact
test) for unpaired data. e analysis was done by the Department of
Biostatistics, Faculty of Health Sciences, UFS, using SAS version 9.4
(SAS, USA).
Results
A total of 182 patients with haematological malignancies were admitted
to the MICU between 1January 2010 and 31 December 2019, of whom
51 (28.0%) fullled the inclusion criteria. Baseline characteristics are
shown in Table1. e median (interquartile range) age was 33 (23 -
49) years, and most of the patients were female (n=29; 56.9%). Most
patients were admitted with either acute myeloid leukaemia (n=22;
43.1%) or acute lymphocytic leukaemia (n=16; 31.4%), while B-cell
lymphoma (n=12; 23.5%) and multiple myeloma (n=1; 2.0%) were less
frequent. Most patients (n=41; 80.4%) had septic shock. e median
APACHE II score for the study population was 23, while the median
AJTCCM VOL. 29 NO. 1 2023 20
RESEARCH
SOFA score was 10. HIV status was under-
reported, with results available for only 26 of
the 51 patients. Of these patients, 13 (50.0%)
were HIV positive. Vasoactive therapy was
required by 44 patients (86.3%), but medical
records for one patient only indicated use
of inotropic support, and not which or how
many agents. Phenylephrine was the most
commonly provided vasopressor (n=40/43;
93.0%), followed by adrenaline (n=18/43;
41.9%), noradrenaline (n=5/43; 11.6%) and
dobutamine (n=3/43; 7.0%). Blood specimens
yielded positive culture results in 31 patients
(60.8%) (9 patients had no growth, 5 cultured
two organisms each, and there were 11 for
whom culture results were not available).
The most frequent organism cultured was
Escherichia coli (n=9/45; 20.0%), followed by
coagulase-negative Staphylococcus (n=7/45;
15.6%), Acinetobacter baumanii (n=6/45;
13.3%) and Klebsiella pneumoniae (n=6/45;
13.3%). No fungi were cultured. Mechanical
ventilation was required by 35/50 patients
(70.0%). e median length of stay in the ICU
was 3 days. ICU mortality was 76.5%, while
hospital mortality (including ICU mortality)
was 82.4%.
ICU and hospital mortality are shown
in Table 2. There were no statistically
signicant age or sex dierences between
survivors and non-survivors. There was
also no statistically significant difference
in mortality rates between the types of
haematological malignancies in patients
admitted to the ICU. e number of patients
with septic shock requiring vasoactive
agents or mechanical ventilation was
statistically higher in the group who did not
survive their ICU stay. Higher APACHE-II
and SOFA scores were also associated with
death. With regard to hospital survival data,
there was a statistically signicantly higher
number of patients in the groups requiring
more than one vasoactive agent and with
high SOFA scores who did not survive.
Table3 shows odds ratios (ORs) for ICU
mortality. e likelihood of death while in
the ICU was increased in patients with septic
shock (OR 4.9), as well as in those requiring
vasoactive agents (OR 13.2) and mechanical
ventilation (OR 8.6).
Discussion
As far as we are aware, this is the rst outcomes
study in patients with haematological
malignancies and febrile neutropenia in a
Table1. Baseline characteristics (N=51)
Variable n (%)*
Age (years), median (IQR) 33 (23 - 49)
Sex
Male 22 (43.1)
Female 29 (56.9)
Type of malignancy
AML 22 (43.1)
ALL 16 (31.4)
B-cell lymphoma 12 (23.5)
Multiple myeloma 1 (2.0)
Reason for admission
Septic shock 41 (80.4)
Sepsis 10 (19.6)
Vasoactive agents
Yes 44 (86.3)
No 7 (13.7)
Types of vasoactive agents (n=43)†
Phenylephrine 40 (93.0)
Adrenaline 18 (41.9)
Noradrenaline 5 (11.6)
Dobutamine 3 (7.0)
Number of vasoactive agents (n=43)
1 22 (51.2)
2 19 (44.2)
3 2 (4.7)
Micro-organisms‡
Escherichia coli 9 (20.0)
Coagulase-negative Staphylococcus 7 (15.6)
Klebsiella pneumoniae 6 (13.3)
Acinetobacter baumannii 6 13.3)
Enterococcus faecium 3 (6.7)
Enterobacter cloacae 1 (2.2)
Flavobacterium orizihabi 1 (2.2)
Methicillin-sensitive Staphylococcus aureus 1 (2.2)
Pseudomonas aeruginosa 1 (2.2)
Methicillin-resistant Staphylococcus aureus 1 (2.2)
No growth 9 (20.0)
Clinical and laboratory results, median (IQR)
GCS score 15 (12 - 15)
Systolic blood pressure (mmHg) (n=50) 101 (93 - 113)
PaO2/FiO2 (n=44) 180 (141 - 258)
White cell count (× 10⁹/L) 0.15 (0.06 - 0.38)
Neutrophil count (× 10⁹/L) 0.03 (0.01 - 0.11)
Platelet count (× 10⁹/L) 15 (10 - 28)
Total bilirubin (× 10⁹/L) 20 (11 - 37)
Creatinine (µg/L) 111 (71 - 212)
Lactate (mmol/L) (n=32) 4.6 (1.9 - 8.1)
HIV-positive status (n=26) 13 (50.0)
Mechanical ventilation (n=50)
Yes 35 (70.0)
No 15 (30.0)
Severity of illness scores, median (IQR)
APACHE-II (n=48) 23 (19 - 29)
SOFA (n=46) 10 (7 - 13)
Outcomes
Length of ICU stay (days), median (IQR) 3 (1 - 5)
ICU mortality 39 (76.5)
Hospital mortality (including ICU mortality) 42 (82.4)
IQR = interquartile range; AML = acute myelogenous leukaemia; ALL = acute lymphoblastic leukaemia; GCS = Glasgow Coma Scale;
PaO2/FiO2 = ratio of arterial oxygen partial pressure to fractional inspired oxygen; APACHE-II = Acute Physiology and Chronic
Health Evaluation; SOFA = Sequential Organ Failure Assessment; ICU = intensive care unit.
*Except where otherwise indicated.
†Some patients received more than one vasoactive agent.
‡Five patients cultured 2 organisms each. ere were 11 patients for whom culture results were not available.
AJTCCM VOL. 29 NO. 1 2023 21
RESEARCH
tertiary hospital ICU in Free State Province, SA. We found an ICU
mortality rate of 76.5% and a hospital mortality rate of 82.4%. ese
results are in stark contrast to ICU and hospital mortality rates of
24.8% and 45.3%, respectively, found by Al-Zubaidi etal.[12] in a
developed-world ICU, and the 38% and 46% reported in a recent
systematic review.[13] e reasons for the high mortality in our ICU
are probably delays in sepsis recognition with a disproportionally high
number of patients with septic shock (80.4%) as opposed to sepsis
only (19.6%),[14] and delays in administering appropriate antibiotic
therapy in patients with septic shock.[15] ICU mortality in our study
was signicantly associated with septic shock and the requirement for
vasoactive agents, as indicated by the ORs in Table3.
Mechanical ventilation was associated with an 8.6-fold increased
ICU mortality in our study. Of the patients who were mechanically
ventilated, 88.6% died in the ICU, as opposed to 46.7% who survived
if mechanical ventilation was not required. In the study by Al-Zubaidi
etal.,[12] mechanically ventilated patients also had higher ICU mortality
of 53.5% and hospital mortality of 75% compared with non-ventilated
patients e gures increased to ICU and hospital mortality of 59.3%
and 90.6%, respectively, for patients who underwent allogeneic
stem cell transplantation. Absence of respiratory failure requiring
mechanical ventilation has been shown to be strongly associated with
survival.[16]
Our study population consisted of young adults, with a median age
of 33 years. e incidence of haematological malignancies other than
Hodgkin’s lymphoma, however, generally tends to increase with age
up to 75 years.[17] Our patients’ young age therefore probably reects
the ICU admission criteria, which prioritise younger patients. ere
Table2. ICU and hospital survival data (N=51)
Variable
ICU survival data, n (%)†Hospital survival data, n (%)†
Survivors
(n=12; 23.6%)
Non-survivors
(n=39; 76.4%) p-value
Survivors
(n=9; 17.6%)
Non-survivors
(n=42; 82.4%) p-value
Age (years), median (IQR) 27 (23.5 - 53.5) 33 (22 - 47) 0.9734 24 (23 - 30) 36 (24 - 50) 0.1192
Sex 0.2242 0.1499
Male (n=22) 7 (31.8) 15 (68.2) 6 (27.3) 16 (72.7)
Female (n=29) 5 (17.2) 24 (82.8) 3 (10.3) 26 (89.7)
Type of malignancy 1.00 1.00
AML (n=22) 5 (22.7) 17 (77.3) 4 (18.2) 18 (81.8)
ALL (n=16) 4 (25.0) 12 (75.0) 3 (18.8) 13 (81.3)
B-cell lymphoma (n=12) 3 (25.0) 9 (75.0) 2 (16.7) 10 (83.3)
Multiple myeloma (n=1) 0 1 (100) 0 1 (100)
Reason for admission 0.0422* 0.3534
Septic shock (n=41) 7 (17.1) 34 (82.9) 6 (14.6) 35 (85.4)
Sepsis (n=10) 5 (50.0) 5 (50.0) 3 (30.0) 7 (70.0)
Vasoactive agents 0.0054* 0.0947
Yes (n=44) 7 (15.9) 37 (84.1) 6 (13.6) 38 (86.4)
No (n=7) 5 (71.4) 2 (28.6) 3 (42.9) 4 (57.1)
Number of vasoactive agents‡0.1459* 0.0458*
1 (n=22) 6 (27.3) 16 (72.7) 6 (27.3) 16 (72.7)
2 (n=19) 1 (5.3) 18 (94.7) 0 19 (100)
3 (n=2) 0 2 (100) 0 2 (100)
Mechanical ventilation§0.0031* 0.1056
Yes (n=35) 4 (11.4) 31 (88.6) 4 (11.4) 31 (88.6)
No (n=15) 8 (53.3) 7 (46.7) 5 (33.3) 10 (66.7)
Severity of illness scores, median (IQR)
APACHE-II (n=48) 20 (16.5 - 22.5) 27 (20 - 30) 0.0207* 20 (18 - 22) 25 (20 - 30) 0.0551
SOFA (n=46) 7 (6 - 8) 11.5 (9 - 13.5) 0.0005* 7 (6 - 8) 11 (9 - 13) 0.0012*
Length of ICU stay (days), median (IQR) 6 (4 - 8) 2 (1 - 5) 0.0065* - - -
ICU = intensive care unit; IQR = interquartile range; AML = acute myelogenous leukaemia; ALL = acute lymphoblastic leukaemia; APACHE-II = Acute Physiology and Chronic Health Evaluation;
SOFA= Sequential Organ Failure Assessment.
*Statistically signicant (p<0.05).
†Except where otherwise indicated.
‡Medical records for one patient indicated use of inotropic support, but not which or how many agents. is patient was therefore not included in this part of the analysis.
§ere was no indication whether one patient was mechanically ventilated or not. is patient was therefore not included in this part of the analysis.
Table3. ORs for ICU mortality
Risk factors OR (95% CI)
Septic shock v. sepsis only 4.9 (1.1 - 21.4)
Vasoactive agents required v. no vasoactive
agents required 13.2 (2.1 - 82.2)
More than one vasoactive agent v. one
vasoactive agent 7.5 (0.8 - 68.8)
Mechanical ventilation v. no mechanical
ventilation 8.6 (2.1 - 37.0)
OR = odds ratio; ICU = intensive care unit; CI = condence interval.
AJTCCM VOL. 29 NO. 1 2023 22
RESEARCH
were no statistically signicant dierences in age between survivors
and non-survivors.
Results on HIV status were only available for 26 of the 51 patients
in our study, and there was a similar number of HIV-positive patients
in the survivor and non-survivor groups. Viral load and CD4 counts
were similarly not available. e inuence of HIV status on mortality
in the era of highly eective antiretroviral therapy is unclear, with
some studies reporting worse outcomes and others equivalent
outcomes.[18,19] In general, however, non-Hodgkin’s lymphoma
remains a common haematological malignancy in people with HIV,
whereas the risk of leukaemia does not seem to be increased.[20] e
risk of high-grade non-Hodgkin’s lymphoma, especially Burkitt’s
lymphoma, is particularly high in HIV-positive patients compared
with HIV-negative patients.[21]
APACHE-II and SOFA scores were higher in non-survivors than
in survivors in our study. is nding is similar to those of other
studies reporting on ICU outcomes of patients with haematological
malignancies.[22,23] Although it is tempting to use illness severity
scores, such as APACHE-II and SOFA, to assist with triage decisions
on individual patients, much more work to improve the accuracy,
validity and predictability of such scoring systems will have to be
conducted before they can be used in clinical practice.[7]
Study limitations
There are several limitations to our study. Firstly, the study was
retrospective and included patient medical records over a 10-year
period. Incomplete medical notes or poor hospital archiving with
lost les may have resulted in missing information. is aspect was
addressed by combining data from hospital paper les and electronic
hospital records to extract the required data.
Secondly, the historically poor survival rates of patients with
haematological malignancies and septic shock could have biased the
decision of the ICU consultant on call whether to admit these patients
to the ICU.
irdly, several confounding factors could have inuenced the
outcomes of the patients. Improved management of haematological
malignancies may have resulted in improved outcomes in general,
with those developing neutropenic fever and septic shock representing
a specic subgroup of patients with higher severity of illness and
increased mortality in the ICU.
e study was conducted in a high HIV prevalence setting,[24] but
the HIV reporting in the study was poor. is could be partially
controlled by reviewing the prescription charts of the patients and
assuming a positive or negative HIV status based on the presence
or absence of antiretroviral agents on the prescription charts.
However, this does not fully control for the lack of HIV data, which
couldpotentially have influenced mortality rates in our study
population.
ICU protocols and clinical management of patients in the ICU
may have changed considerably over 10 years. is may have resulted
in better outcomes in patients admitted to the ICU more recently.
Alternatively, an increased prevalence of multidrug-resistant
pathogens in the hospital and ICU environment may have resulted
in a worse outcome in patients admitted to the ICU more recently.
None of these variables could be denitively accounted for. Finally, the
timeous management and institution of appropriate antibiotic therapy
for septic shock at the time of recognition before transfer to the ICU
could have inuenced ICU and hospital survival rates.
Conclusion
Patients with haematological malignancies and febrile neutropenia in
the UAH MICU in Free State Province have high ICU and hospital
mortality rates. Mortality is associated with septic shock and vasoactive
agent use, mechanical ventilation and high APACHE-II and SOFA
scores. e study was conducted in a resource-limited setting with
strict ICU admission criteria and may therefore not be generalisable to
well-resourced healthcare settings. More needs to be done with regard
to timeous management of patients with haematological malignancies
and septic shock in our setting to improve ICU survival. Strict
infection prevention and control measures should be implemented
in all haematology wards. ese measures need to be adhered to
vigilantly. Tools such as early warning and quick SOFA scores can be
used to recognise and implement treatment for sepsis early. Clinicians
working in haematology wards should be trained to manage patients
with sepsis and septic shock according to the Surviving Sepsis
Campaign (SSC) guidelines. Timeous management of patients with
haematological malignancies and neutropenic sepsis could result in
improved ICU and hospital survival rates.
Declaration. SDM is a member of the editorial board.
Acknowledgements. We thank Mr Cornel van Rooyen, biostatistician,
Department of Biostatistics, Faculty of Health Sciences, University of the
Free State, for statistical analysis of the data, and Ms T Mulder, medical
editor/writer, Faculty of Health Sciences, University of the Free State, for
technical and editorial preparation of the manuscript.
Author contributions. Both CDSM and SDM contributed equally to the
conceptualisation and design of the study, collected the data and wrote
the rst dra of the manuscript. ey also interpreted and discussed the
results and approved the nal dra.
Funding. None.
Conicts of interest.None.
1. World Health Organization. Cancer Today. Estimated number of new cases in 2020, Africa,
South Africa, both sexes, all ages (excl. NMSC). 2020. http://gco.iarc.fr/today/online-
analysis-table?v=2018&mode=cancer&mode_population=countries&population
=900&populations=903_710&key=asr&sex=0&cancer=39&type=0&statistic
=5&prevalence=0&population_group=0&ages_group%5B%5D=0&ages_
group%5B%5D=17&nb_items=5&gro (accessed 31 March 2022).
2. Taccone FS, Artigas AA, Sprung CL, Moreno R, Sakr Y, Vincent JL. Characteristics
and outcomes of cancer patients in European ICUs. Crit Care 2009;13(1):R15. https://
doi.org/10.1186/cc7713
3. Darmon M, Bourmaud A, Georges Q, et al. Changes in critically ill cancer patients’
short-term outcome over the last decades: Results of systematic review with meta-
analysis on individual data. Intensive Care Med 2019;45(7):977-987. https://doi.
org/10.1007/s00134-019-05653-7
4. Azoulay E, Mokart D, Pène F, et al. Outcomes of critically ill patients with hematologic
malignancies: Prospective multicenter data from France and Belgium – a Groupe
de Recherche Respiratoire en Réanimation Onco-hématologique study. J Clin Oncol
2013;31(22):2810-2818. https://doi.org/10.1200/JCO.2012.47.2365
5. Turki AT, Liebregts T. How I treat: Why patients with hematological malignancies
should be admitted to an intensive care unit. Biomed J Sci Tech Res 2018;2(3):2699-
2700. https://doi.org/10.26717/BJSTR.2018.02.000774
6. Azoulay E, Pène F, Darmon M, et al. Managing critically ill hematology patients:
Time to think dierently. Blood Rev 2015;29(6):359-367. https://doi.org/10.1016/j.
blre.2015.04.002
7. Joynt GM, Gopalan DP, Argent AA, et al. e Critical Care Society of Southern
Africa consensus statement on ICU triage and rationing (ConICTri). S Afr Med J
2019;109(8b):613-629. https://doi.org/10.7196/SAMJ.2019.v109i8b.13947
AJTCCM VOL. 29 NO. 1 2023 23
RESEARCH
8. Bird GT, Farquhar-Smith P, Wigmore T, Potter M, Gruber PC. Outcomes and
prognostic factors in patients with haematological malignancy admitted to a specialist
cancer intensive care unit: A 5 yr study. Br J Anaesth 2012;108(3):452-459. https://doi.
org/10.1093/bja/aer449
9. Grgić Medić M, Gornik I, Gašparović V. Hematologic malignancies in the medical
intensive care unit – outcomes and prognostic factors. Hematology 2015;20(5):247-253.
https://doi.org/10.1179/1607845414Y.0000000206
10. Rawson JL, Fagan FM, Burrough GC, et al. Intensive care unit outcomes in patients
with hematological malignancy. Blood Sci 2020;2(1):33-37. https://doi.org/10.1097/
BS9.0000000000000038
11. Probst L, Schalk E, Liebregts T, et al. Prognostic accuracy of SOFA, qSOFA and SIRS
criteria in hematological cancer patients: A retrospective multicenter study. J Intensive
Care 2019;7:41. https://doi.org/10.1186/s40560-019-0396-y
12. Al-Zubaidi N, Shehada E, Alshabani K, ZazaDitYafawi J, Kingah P, Soubani AO.
Predictors of outcome in patients with hematologic malignancies admitted to the
intensive care unit. Hematol Oncol Stem Cell er 2018;11(4):206-218. https://doi.
org/10.1016/j.hemonc.2018.03.003
13. Nazer L, Brown AR, Cuenca J, et al. 483: A systematic review evaluating mortality in
patients with cancer admitted to the intensive care unit. Crit Care Med 2021;49(1):232.
https://doi.org/10.1097/01.ccm.0000727820.04292.45
14. Bauer M, Gerlach H, Vogelmann T, Preissing F, Stiefel J, Adam D. Mortality in sepsis
and septic shock in Europe, North America and Australia between 2009 and 2019 –
results from a systematic review and meta-analysis. Crit Care 2020;24(1):239. https://
doi.org/10.1186/s13054-020-02950-2
15. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation
of eective antimicrobial therapy is the critical determinant of survival in human
septic shock. Crit Care Med 2006;34(6):1589-1596. https://doi.org/10.1097/01.
CCM.0000217961.75225.E9
16. De Vries VA, Müller MCA, Arbous MS, et al. Long-term outcome of patients with a
hematologic malignancy and multiple organ failure admitted at the intensive care. Crit
Care Med 2019;47(2):e120-e128. https://doi.org/10.1097/CCM.0000000000003526
17. Schonfeld SJ, Erdmann F, Wiggill T, et al. Hematologic malignancies in South Africa
2000 - 2006: Analysis of data reported to the National Cancer Registry. Cancer Med
2016;5(4):728-738. https://doi.org/10.1002/cam4.597
18. Cingolani A, Cozzi Lepri A, Teoli L, et al. Survival and predictors of death in people
with HIV-associated lymphoma compared to those with a diagnosis of lymphoma
in general population. PLoS ONE 2017;12(10):e0186549. https://doi.org/10.1371/
journal.pone.0186549
19. Moahi K, Ralefala T, Nkele I, et al. HIV and Hodgkin lymphoma survival: A
prospective study in Botswana. JCO Glob Oncol 2022;8:e2100163. https://doi.
org/10.1200/GO.21.00163
20. Kimani SM, Painschab MS, Horner MJ, et al. Epidemiology of haematological
malignancies in people living with HIV. Lancet HIV 2020;7(9):e641-e651. https://
doi.org/10.1016/S2352-3018(20)30118-1
21. Dhokotera T, Bohlius J, Spoerri A, et al. e burden of cancers associated with HIV
in the South African public health sector, 2004 - 2014: A record linkage study. Infect
Agent Cancer 2019;14:12. https://doi.org/10.1186/s13027-019-0228-7
22. Alp E, Tok T, Kaynar L, et al. Outcomes for haematological cancer patients admitted
to an intensive care unit in a university hospital. Aust Crit Care 2018;31(6):363-368.
https://doi.org/10.1016/j.aucc.2017.10.005
23. Cetintepe T, Cetintepe L, Solmaz S, et al. Determination of the relationship between
mortality and SOFA, qSOFA, MASCC scores in febrile neutropenic patients monitored
in the intensive care unit. Support Care Cancer 2021;29(7):4089-4094. https://doi.
org/10.1007/s00520-020-05924-w
24. World Health Organization. South Africa. HIV country prole. 2021. https://cfs.hivci.
org/ (accessed 19 May 2022).
Accepted 17 January 2023.
