48 AJTCCM VOL. 29 NO. 2 2023
Background. Most asthma-related deaths occur in low- and middle-income countries, and South Africa (SA) is ranked h in global asthma
mortality. Little is known about the characteristics and outcome of asthma patients requiring intensive care unit (ICU) admission in SA.
Objectives. To identify and characterise patients with acute severe asthma admitted to the respiratory ICU at Groote Schuur Hospital, Cape
Town, SA, in order to evaluate outcomes and identify predictors of poor outcomes in those admitted.
Methods. We performed a retrospective descriptive study of patients with severe asthma admitted to the respiratory ICU at Groote Schuur
Hospital between 1 January 2014 and 31 December 2019.
Results. One hundred and three patients (110 admission episodes) were identied with an acute asthma exacerbation requiring ICU
admission; all were mechanically ventilated. ere was a female preponderance (53.6%; n=59/110), with a median (range) age overall of
33 (13 - 84) years. Of all admissions, 40 (36.4%) were current tobacco smokers and 16 (14.5%) patients with a history of substance abuse.
Twothirds (60.0%; n=66/110) of the patients were using an inhaled corticosteroid (ICS). No predictors of mortality were evident in
multivariate modelling, although those who died were older, and had higher Acute Physiology and Chronic Health Evaluation (APACHEII)
scores and longer duration of admission. Only 59 of the surviving 96 individual patients (61.5%) attended a specialist pulmonology clinic
aer discharge.
Conclusion. Among patients admitted to the respiratory ICU at Groote Schuur Hospital for asthma exacerbations, there was a high
prevalence of smokers and poor coverage with inhaled ICSs. Although mortality was low compared with general ICU mortality, more
needs to be done to prevent acute severe asthma exacerbations.
Keywords. Asthma, intensive care, outcomes.
Afr J Thoracic Crit Care Med 2023;29(2):e212. https://doi.org/10.7196/AJTCCM.2023.v29i2.212
Asthma is one of the most prevalent chronic diseases globally,
with an estimated 339 million people diagnosed with asthma. A
disproportionately high number of asthma-related deaths occur in low-
and middle-income countries (LMICs); according to the World Health
Organization, >80% of asthma-related deaths occur in these countries.[1]
Factors that have been attributed to the high asthma-related mortality
in LMICs include lack of access to essential medicines, suboptimal long-
term medical care, delays in obtaining medical care, low income, and
low levels of education.[2,3] Hospital admissions can be seen as an indirect
measure of the burden of uncontrolled and more severe asthma and lack
of eective primary asthma care. Treatment in the intensive care unit
(ICU), especially mechanical ventilation, represents the most severely
aected patients and those who have not been adequately controlled in
the outpatient setting.[4,5]
There is no direct correlation between asthma-related mortality
and asthma prevalence. South Africa (SA) is currently ranked 25th
Asthma in the intensive care unit: A review of patient
characteristics and outcomes
F Manyeruke,1 MB BCh, MMed (Med), Cert Pulm (SA); G L Calligaro,1 BSc Hons, MB BCh, Dip PEC (SA), FCP (SA), MMed (Med), Cert
Pulm (SA); R Raine,1 MB ChB, MMed (Med), FCP (SA);
R N van Zyl-Smit,1 MB ChB, FRCP (UK), FCP (SA), MMed (Med), Dip HIV Man (SA), Cert Pulm (SA), PhD
Division of Pulmonology and UCT Lung Institute, Department of Medicine, Faculty of Health Sciences, University of Cape Town and Groote Schuur Hospital,
Cape Town, South Africa
Corresponding author: R N van Zyl Smit (richard.vanzyl-smit@uct.ac.za)
Study synopsis
What the study adds. Intensive care unit (ICU) admission represents the most severe form of exacerbation of asthma. South Africa (SA)
has a very high rate of asthma deaths, and this study demonstrates that admission to an ICU with a very severe asthma exacerbation
frequently results in a good outcome. However, many of the patients admitted to the ICU were not adequately treated with background
asthma medications prior to their admission.
Implications of the ndings. Death from asthma should be avoidable, and admission to an ICU is not associated with high mortality.
Patients are therefore likely to be dying at home or out of hospital. Better education and access to medication and early access to health
services rather than improved in-hospital care would potentially alter SA’s high asthma mortality.
ORIGINAL RESEARCH: ARTICLES
AJTCCM VOL. 29 NO. 2 2023 49
worldwide for asthma prevalence, but 5th in global asthma mortality.[5]
However, not all people with severe asthma will make it to an ICU;the
majority (>70%) of asthma deaths in SA occur before arrival at a
hospital.[6]
We therefore set out to conduct a retrospective descriptive study to
identify and characterise all patients with acute severe asthma admitted
to the respiratory ICU at Groote Schuur Hospital, Cape Town, SA, in
order to evaluate outcomes and identify predictors of poor outcomes
in those admitted.
Methods
We performed a retrospective descriptive study of patients with acute
severe asthma admitted to the respiratory ICU at Groote Schuur Hospital.
The medical/respiratory ICU is an 8-bed unit admitting primarily
medical patients requiring intensive care and/or mechanical ventilation.
e records of patients admitted to the ICU between 1January 2014 and
31 December 2019 (prior to the COVID -19 pandemic) were reviewed,
and patients admitted with a primary diagnosis of acute severe asthma
were included in the study.
Demographic data, blood results on admission, arterial blood gas
measurements, Acute Physiology and Chronic Health Evaluation
(APACHE II) score, comorbidities, current medications, length of ICU
stay, complications in the ICU, days mechanically ventilated, duration
of hospital stay, and outcomes were collected. All data were entered into
a REDCap (Research Electronic Data Capture) database.
Statistical analysis
e data were analysed using GraphPad Prism 9 for macOS (GraphPad
Software LLC, USA). Descriptive statistics and Fisher’s exact tests
were performed on dichotomous categorical variables and t-tests on
continuous data. Parametric and non-parametric tests were used in
univariate analyses. Predictors of outcome were explored using stepwise
multivariate logistic regression analysis. A signicance level of p<0.05
was chosen, and all statistical tests were two-sided.
Ethical considerations
The University of Cape Town Faculty of Health Sciences Human
Research Ethics Committee (ref. no. HREC REF:110/2020) and the
Groote Schuur Hospital Research Committee provided approval to
conduct the audit.
Results
During the 6 years 2014 - 2019, there were 2 684 admissions to the
respiratory ICU. Of these, a total of 110 individual admissions (103
patients) had a diagnosis of an acute severe asthma exacerbation. ese
constituted 4.1% of the total ICU admissions. All patients required
mechanical ventilation. ere was a slight female preponderance (53.6%;
n=59/110). e patients were generally young, with a median (range)
age of 33.5 (13 - 84) years (Fig.1). e demographic characteristics
(individual admissions) are detailed in Table1.
Over a third (36.4%; n=40/110 individual admissions) of the patients
were current tobacco smokers, and 14.5% (n=16/110) had a history of
substance abuse; 93.8% of substance abusers smoked tobacco, compared
with 26.7% of patients with no history of substance abuse (p<0.001).
Patients with a history of substance abuse were signicantly younger
than the non-substance users (29 years v. 37.9 years; p=0.002).
Almost all (91.8%; n=101/110 individual admissions) of the patients
were using a short-acting beta-agonist, but only 60.0% (n=66/110)
reported being treated with an inhaled corticosteroid (ICS) before the
ICU admission. Half of those on an ICS were additionally receiving a
long-acting beta-agonist, either as a separate inhaler ‘free’ or in xed-
dose combination with an inhaled steroid. The prevalence of ICS
treatment was signicantly lower (25.0%) in patients with a history
of substance abuse compared with patients without such a history
(69.9%) (p=0.004). A small proportion of patients (10.9%; n=12/110)
were on long-term oral corticosteroids. e frequency of prescribed
asthma medications taken by the patients admitted to the ICU is shown
in Table2.
All patients were intubated prior to transfer to the ICU. On admission
to the ICU, 82.7% (n=91/110 individual admissions) had acute
respiratory acidosis, median (interquartile range (IQR)) pH 7.22 (7.13
- 7.31), and 10.9% had a pH <7.0. e median partial pressure of carbon
dioxide (PaCO2) was 9.09 (6.68 - 11.25) kPa, and 85.5% (n=94/110) of
the patients had signicant hypercapnia. e median white cell count
was 15.30 (12.30 - 22.20) × 109/L, with low lactate levels (1.30 (0.90 -
1.93) mmol/L) (Table3).
e median (IQR) length of mechanical ventilation was 2 (1 - 4) days.
ICU stay aer extubation was short, with a median overall ICU stay of
3 (2 - 5) days. A minority of the patients (10.0%; n=11/110 individual
admissions) had a prolonged ICU stay (>7days). e median length of
total stay in hospital was 5 (4 - 8) days (Fig.2).
Less than a quarter of the patients (22.7%; n=25/110 individual
admissions) had a recorded complication during their ICU stay. e
most common complications were renal failure (13.6%; n=15/110),
sepsis (10.9%; n=12/110), pneumothorax (2.7%; n=3/110) and cardiac
arrhythmias (2.7%; n=3/110). Very few patients had more than one
complication.
e overall mortality in the ICU (individual admissions) was 4.5%
(n=5 deaths); 103 patients (93.6%) survived to hospital discharge, with
2 deaths occurring aer ICU discharge. Patients who died were on
average older and had a longer length of stay and higher APACHEII
scores compared with those who survived to ICU discharge (n=105)
(Table4). No predictors of mortality were evident in the multivariate
analysis.
Among the 96 individual patients who were discharged from hospital
20
15
10
5
0
Relative frequency, %
Age (in 5-year bands)
15 20 25 30 35 40 45 50 55 60 65 70 75 80 85
Fig.1. Age distribution of patients admitted with severe asthma for
ventilation. (N=110 individual admissions.)
ORIGINAL RESEARCH: ARTICLES
50 AJTCCM VOL. 29 NO. 2 2023
aer their ICU admission (7 of the 103 had died), 59 (61.5%) had a record
of a subsequent follow-up at the specialist respiratory clinic at Groote
Schuur Hospital. A smaller percentage (25.0%; n=4/16) of patients with
a history of substance abuse returned for follow-up compared with non-
substance users (68.8%; n=55/80) (p=0.01). e odds of returning for
review aer discharge among patients who had a history of substance
abuse was 0.21 (95% condence interval 0.07 - 0.69).
Discussion
We present the outcomes and clinical characteristics of the 2014 - 2019
admissions to the Groote Schuur Hospital respiratory ICU of patients
with acute severe asthma requiring invasive mechanical ventilation.
Ventilation for asthma represents <5% of ICU admissions, with >95%
survival. However, ICU mortality for asthma at our institution has
eectively doubled from the previously documented gure of 2.2%
for 1980 - 1997.[6] It is unclear why mortality has increased, given the
advances in ICU care; however, it remains comparable to high-income
countries, which have an ICU mortality for mechanically ventilated
asthmatics of ~5%.[4,7]
ICU admission is synonymous with failure of preventive treatment
in the healthcare system to prevent severe exacerbations of asthma. e
younger age and female preponderance were seen in a previous asthma
ICU study in SA, with similar (4.3 days) ICU stay.[6] Long delays in
seeking medical attention and underuse of corticosteroids were risk
factors for invasive mechanical ventilation.[8]
Only two-thirds of the patients admitted with acute severe asthma in
the present study reported being treated with an ICS. Information with
regard to adherence and inhaler technique was not obtainable. is low
use of ICSs among asthmatics is of concern, given the association with an
increased risk of hospitalisation and mortality[9-11] as well as availability
of the Global Initiative for Asthma (GINA) strategy document and SA
asthma guidelines, which both reinforce the need for all asthma patients
to receive an ICS.[5]
Globally it has been reported that the major risk factors for ICU
admission for acute severe asthma include a history of smoking
(>10pack-years), atopy, previous mechanical ventilation, and high
use of an inhaled beta-agonist.[12-14] Of the patients in the present study
(individual admissions), 36.4% were smokers. is gure is higher
than the background estimated population smoking rates of 32.9% in
Western Cape Province and 17.6% in SA.[15] Asthmacontrol,response
to steroids and outcomes are worse in smokers and those abusing other
drugs,[16-18] but in our cohort, all the asthmatics who smoked and/or
used illicit drugs survived. However, return for follow-up was worse in
those who smoked and reported drug use.
ere were several limitations to this study. Owing to its retrospective
nature and our reliance on medical records, it was not possible to obtain
accurate data on medication history and smoking rates, and other
information. Specically, details on the amount of rescue medicine
used prior to admission, and adherence to using ICSs, are not available.
Table1. Demographics and baseline clinical information on
patients admitted to the ICU with asthma (N=110 individual
admissions)
Variabl e n (%)*
Age (years), median (range) 33.5 (13 - 84)
Male 51 (46.4)
Female 59 (53.6)
Previous ICU admission 7 (6.4)
Current tobacco smoker 40 (36.4)
Hypertension 16 (14.5)
Substance abuse 16 (14.5)
Diabetic 10 (9.1)
HIV 6 (5.4)
Pregnant 5 (4.5)
ICU = intensive care unit.
*Except where otherwise indicated.
Table2. Asthma-related medication documented on
admission to the ICU (N=110 individual admissions)
Medication n (%)
SABA 101 (91.8)
ICS 66 (60.0)
LABA* 32 (29.1)
SAMA 13 (11.8)
OCS 12 (10.9)
Oral theophylline 24 (21.8)
LAMA 2 (1.8)
ICU = intensive care unit; SABA = short-acting beta-agonist; ICS = inhaled corticosteroid;
LABA = long-acting beta-agonist; OCS = oral corticosteroid; LAMA = long-acting muscarinic
antagonist.
*All patients receiving a LABA were also co-prescribed an ICS.
Table3. Blood results on admission to the ICU (N=110
individual admissions)
Variabl e Median (IQR)
pH 7.22 (7.13 - 7.31)
PaO2 (kPa) 13.40 (10.73 - 18.15)
PaCO2 (kPa) 9.09 (6.68 - 11.25)
HCO3 (mmol/L) 24.15 (21.0 - 28.43)
WCC (× 109/L) 15.30 (12.30 - 22.20)
Hb (g/dL) 12.40 (11.30 - 13.80)
Lactate (mmol/L) 1.30 (0.90 - 1.93)
Creatinine (µmol/L) 64.50 (52.0 - 84.75)
APACHE II score 9 (8 - 12)
ICU = intensive care unit; IQR = interquartile range; PaO2 = partial pressure of oxygen; P
aCO2 = partial pressure of carbon dioxide; HCO3 = bicarbonate; WCC = white cell count;
Hb = haemoglobin; APACHE II = Acute Physiology and Chronic Health Evaluation.
50
40
30
20
10
0
Patients, %
2 4 6 8 10 12 14 16 18 20 22 24 26 28
Days in ICU
Fig.2. Length of stay of asthma patients admitted to the ICU. (ICU =
intensive care unit.)
ORIGINAL RESEARCH: ARTICLES
AJTCCM VOL. 29 NO. 2 2023 51
The potential exists that some patients with chronic obstructive
pulmonary disease (COPD) may have been included in the acute severe
asthma cohort, and that some patients with acute severe asthma were
erroneously labelled as having COPD. Without complete follow-up of
both groups, it is not possible to be 100% certain.
Conclusion
Approximately 18 patients with acute severe asthma are admitted per year
to the Groote Schuur Hospital ICU for mechanical ventilation, a third of
whom are not being treated with an ICS. ese patients frequently smoke
and use drugs. More work is required to identify patients in primary
care who are at risk of exacerbations and ensure that optimal therapy is
provided to prevent ICU admission for mechanical ventilation.
Declaration. GLC, RR and RvZ-S are members of the editorial board. e
research for this study was done in partial fullment of the requirements for
FM’s Cert Pulm (SA) degree at the University of Cape Town, as stipulated
by the Health Professions Council of South Africa and the South African
oracic Society.
Acknowledgements. We are grateful to the sta in the Division of Critical
Care at Groote Schuur Hospital who cared for these patients and allowed us
to evaluate the outcomes.
Author contributions. RR, RNvZ-S and FM conceptualised the study, FM
collected the data, and RNvZ-S analysed the data. All authors contributed
to data review and manuscript preparation.
Funding.None.
Conicts of interest.None.
1. Brakema EA, Tabyshova A, van der Kleij R, etal. e socioeconomic burden of chronic
lung disease in low-resource settings across the globe – an observational FRESH AIR
study. Respir Res 2019;20:291. https://doi.org/10.1186/s12931-019-1255-z
2. Sinharoy A, Mitra S, Mondal P. Socioeconomic and environmental predictors of
asthma-related mortality. J Environ Public Health 2018;2018:9389570. https://doi.
org/10.1155/2018/9389570
3. Grant EN, Lyttle CS, Weiss KB. e relation of socioeconomic factors and racial/ethnic
dierences in US asthma mortality. Am J Public Health 2000;90(12):1923-1925. https://
doi.org/10.2105/ajph.90.12.1923
4. Gibbison B, Griggs K, Mukherjee M, Sheikh A. Ten years of asthma admissions to adult
critical care units in England and Wales. BMJ Open 2013;3(9):e003420. https://doi.
org/10.1136/bmjopen-2013-003420
5. Global Initiative for Asthma. Global Strategy for Asthma Management and
Prevention, 2021. https://ginasthma.org/wp-content/uploads/2023/04/GINA-Main-
Report-2021-V2-WMSA.pdf (accessed August 2021).
6. Zar HJ, Stickells D, Toerien A, Wilson D, Klein M, Bateman ED. Changes in fatal
and near-fatal asthma in an urban area of South Africa from 1980 - 1997. Eur Respir
J 2001;18(1):33-37. https://doi.org/10.1183/09031936.01.00081801
7. Demoule A, Brochard L, Dres M, etal. How to ventilate obstructive and asthmatic
patients. Intensive Care Med 2020;46(12):2436-2449. https://doi.org/10.1007/
s00134-020-06291-0
8. Westerman DE, Benatar SR, Potgieter PD, Ferguson AD. Identication of the high-
risk asthmatic patient: Experience with 39 patients undergoing ventilation for
status asthmaticus. Am J Med 1979;66(4):565-572. https://doi.org/10.1016/0002-
9343(79)91165-3
9. Gonem S, Cumella A, Richardson M. Asthma admission rates and patterns of
salbutamol and inhaled corticosteroid prescribing in England from 2013 to 2017.
orax 2019;74(7):705-706. https://doi.org/10.1136/thoraxjnl-2018-212723
10. Weatherall M, Wijesinghe M, Perrin K, Harwood M, Beasley R. Meta-analysis of the
risk of mortality with salmeterol and the eect of concomitant inhaled corticosteroid
therapy. orax 2010;65(1):39-43. https://doi.org/10.1136/thx.2009.116608
11. O’Byrne P, Fabbri LM, Pavord ID, Papi A, Petruzzelli S, Lange P. Asthma progression
and mortality: e role of inhaled corticosteroids. Eur Respir J 2019;54(1):1900491.
https://doi.org/10.1183/13993003.00491-2019
12. Juárez León JE, Gonzalez Islas D, Sandoval Gutierrez JL, Miguel Reyes JL, Orea
Tejeda A. Risk factors for hospitalisation and ICU admission for near-fatal
asthma. Eur Respir J 2017;50(suppl 61):PA3575. https://doi.org/10.1183/1393003.
congress-2017.PA3575
13. Eisner MD, Lieu TA, Chi F, etal. Beta agonists, inhaled steroids, and the risk of
intensive care unit admission for asthma. Eur Respir J 2001;17(2):233-240. https://
doi.org/10.1183/09031936.01.17202330
14. Nakwan N, Prateepchaiboon T. Predicting the requiring intubation and invasive
mechanical ventilation among asthmatic exacerbation-related hospitalisations. J
Asthma 2022;59(3):507-513. https://doi.org/10.1080/02770903.2020.1853768
15. Reddy P, Zuma K, Shisana O, Kim J, Sewpaul R. Prevalence of tobacco use among
adults in South Africa: Results from the rst South African National Health and
Nutrition Examination Survey. S Afr Med J 2015;105(8):648-655. https://doi.
org/10.7196/SAMJnew.7932
16. Levine M, Iliescu ME, Margellos-Anast H, Estarziau M, Ansell DA. e eects of
cocaine and heroin use on intubation rates and hospital utilisation in patients with
acute asthma exacerbations. Chest 2005;128(4):1951-1957. https://doi.org/10.1378/
chest.128.4.1951
17. Polosa R, omson NC. Smoking and asthma: Dangerous liaisons. Eur Respir J
2013;41(3):716-726. https://doi.org/10.1183/09031936.00073312
18. Caponnetto P, Auditore R, Russo C, etal. ‘Dangerous relationships’: Asthma and
substance abuse. J Addict Dis 2013;32(2):158-167. https://doi.org/10.1080/10550
887.2013.795469
Received 28 February 2022. Accepted 2 May 2023. Published xx June 2023.
Table4. Univariate analysis of factors associated with mortality
Variabl e ICU survivors (n=105) ICU non-survivors (n=5) p-value OR (univariate analysis)
pH, mean (SD) 7.21 (0.14) 7.13 (0.26) 0.165 -
PaCO2 (kPa), mean (SD) 9.36 (3.81) 11.10 (6.34) 0.268 -
Age (years), mean (SD) 35.6 (12.4) 51.3 (23.8) 0.003 0.94 (0.88 - 0.99)
APACHE II score, mean (SD) 9.8 (3.7) 14.4 (3.5) 0.002 0.84 (0.70 - 1.02)
Days in ICU, mean (SD) 3.7 (2.5) 7.1 (9.3) 0.008 0.87 (0.73 - 1.05)
Smokers, n (%)* 40 (38.1) 0 0.15 -
Reported illicit drug use, n (%)* 16 (15.2) 0 0.9 -
ICU = intensive care unit; OR = odds ratio; SD = standard deviation; PaCO2 = partial pressure of carbon dioxide; APACHE II = Acute Physiology and Chronic Health Evaluation.
*Of 105 ICU admissions (including readmissions); surviving patients with documentation of substance use/non-use.
ORIGINAL RESEARCH: ARTICLES
