152 AJTCCM VOL. 30 NO. 4 2024
ORIGINAL RESEARCH: ARTICLES
Background. Induced sputum (IS) is a sampling technique for obtaining lower airway samples for microbial investigations, including
GeneXpert and culture for microbiological conrmation of Mycobacterium tuberculosis.
Objectives. To investigate the safety and yield of IS in children admitted to a tertiary hospital in Ghana with presumed pulmonary
tuberculosis (PTB).
Methods. A prospective cross-sectional study was carried out in children aged 3 months - 14 years at Komfo Anokye Teaching Hospital in
Kumasi, Ghana, over the 6-month period January - June 2022. All children with breathing diculty and other signs of respiratory distress
were given respiratory support, and IS samples were obtained when respiratory distress had resolved. One or two IS samples were collected
from each child within 48 hours of admission by a trained nurse aer at least 4 hours of fasting. Children were monitored during and for
30 minutes aer the procedure, with recording of respiratory rate, oxygen saturation, temperature and pulse rate.ey were also monitored
for any adverse events such as vomiting, wheezing and nosebleeds.
Results. A total of 144 children were sampled, with approximately two-thirds sampled a second time. Nearly half of the participants were
aged <2 years (49.3%; n=71/144), and the median (interquartile range (IQR)) age was 2.5 (0.9 - 6.8) years. Ninety-eight children (68.1%)
tested positive for PTB by Xpert Ultra, with 19/98 (19.4%) being rifampicin resistant; 47/102 (46.1%) were positive by Ziehl-Neelsen staining,
and 57/102 (55.9%) were positive by Auramine O staining. ree children (2.1%) had an episode of epistaxis following the procedure. No
other adverse events were observed. Measurements before and 30 minutes to 1 hour aer the procedure (median (IQR)) were similar:
temperature 36.5oC (36.5 - 37.5oC) v. 36.5oC (36.2 - 37.1oC), oxygen saturation 98% (92 - 99%) v. 98% (93 - 99%), pulse rate 120 (106 - 139)
v. 125 (112 - 142) bpm, and respiratory rate 38 (30 - 48) v. 33 (30 - 45) cycles per minute.
Conclusion. We found sputum induction to be a safe and well-tolerated procedure in the paediatric population, with minimal clinical risk
and a high microbiological yield for PTB.
Keywords. Children, tuberculosis, induced sputum.
Afr J Thoracic Crit Care Med 2024;30(4):e1841. https://doi.org/10.7196/AJTCCM.2024.v30i4.1841
Study synopsis
What the study adds. is study is the rst to provide information on the performance and safety of induced sputum (IS) in Ghanaian
children. It shows that IS can be performed safely in this population, despite safety concerns that resulted in its late introduction in the
country. In addition, it shows that IS procedures can provide quality sputum samples to improve bacteriological conrmation of pulmonary
tuberculosis (PTB) in children with presumed tuberculosis. Lastly, it adds to the existing body of literature showing that with requisite
training, sputum induction can be performed in low-income settings.
Implications of the ndings. e study shows that the IS procedure can be used in Ghana to help shi from traditional ways of obtaining
sputum samples in children, such as gastric lavage and routine methods of obtaining sputum in older children without induction, to improve
bacteriological conrmation when PTB is suspected. e ndings indicate that roll-out to other health facilities in Ghana is possible.
Safety and yield of sputum induction for diagnosis of pulmonary
tuberculosis in children in a tertiary hospital in Ghana
K S Owusu,1 BSc, MB ChB, MPhil ; S Kwarteng-Owusu,2,3 BSc, MB ChB, MPhil ; N Wireko-Brobby,2,3 BSc, MB ChB, MSc ;
E Osei,2 BSc; E Abra,2 BSc; F A Appiah,2 BSc; A Enimil,2,3 BSc, MB ChB, MPhil ; J Sylverken,2,3 MD ; A Owusu-Ofori,1 PhD ;
D M Gray,4 FC Paed (SA), PhD ; D Ansong,2,3 BSc, MB ChB, MSc ; H J Zar,5 PhD
1 Department of Clinical Microbiology, School of Medicine and Dentistry, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
2 Directorate of Child Health, Komfo Anokye Teaching Hospital, Kumasi, Ghana
3 Department of Child Health, School of Medicine and Dentistry, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
4 Division of Paediatric Pulmonology, Department of Child Health, Faculty of Health Sciences, University of Cape Town, South Africa
5 Department of Paediatrics and Child Health and MRC Unit on Child and Adolescent Health, University of Cape Town and Red Cross War Memorial Children’s
Hospital, Cape Town, South Africa
Corresponding author: S Kwarteng-Owusu (abenaboamah18@gmail.com, sandra.kwarteng@knust.edu.gh)
AJTCCM VOL. 30 NO. 4 2024 153
ORIGINAL RESEARCH: ARTICLES
Tuberculosis (TB) is one of the leading causes of death and morbidity
in young children, with a recent Global Tuberculosis Report showing
that 12% of TB cases were in children in 2023.[1] Children with TB
may progress to severe disease and death, but the majority of cases
are undiagnosed. One key factor resulting in poor childhood TB
case ascertainment is the difficulty of microbial confirmation in
children. e main challenge is obtaining suitable respiratory samples,
particularly sputum, as young children do not expectorate easily. In
the few cases where expectorated sputum samples are obtained, an
additional problem is that samples may be of poor quality and low in
volume, with bacillary concentrations below the detection threshold
of the diagnostic test.[2]
Induced sputum (IS) is the method of obtaining samples that
provides the highest sensitivity for the detection of Mycobacterium
tuberculosis (MTB) using new rapid polymerase chain reaction
methods, especially Xpert MTB/RIF Ultra.[3]
IS has been shown to be feasible and eective in several low- to
middle-income countries (LMICs).[4-6] It involves a minimum of 4
hours’ fasting, nebulisation with 3 - 5% hypertonic saline, and inhaled
salbutamol. Children are then encouraged to cough sputum up, or
sputum is suctioned through the nasopharynx to obtain samples
if they are unable to expectorate. Data show that the procedure is
generally well tolerated and safe.[4,5] Reported side-eects include
epistaxis, vomiting, wheezing, oxygen desaturation and cough.[6-8]
e procedure can be performed by trained healthcare workers and
is suitable for low-resource facilities. An additional advantage of IS
is that, unlike gastric lavage, it does not require an overnight fast or
hospitalisation, so it can be used more easily in high-burden LMICs.[7-9]
Despite the documented advantages of IS, there are no reports on
the safety and usefulness of the procedure in Ghanaian children.[8,9]
is study aimed to investigate the safety and yield of IS in children
presenting to hospital with suspected TB disease.
Methods
A prospective cross-sectional study was conducted over the 6-month
period January - June 2022 at Komfo Anokye Teaching Hospital
(KATH), a tertiary hospital and major referral centre in Kumasi, the
capital city of the Ashanti region of Ghana.
Study participants
e study participants were children in the paediatrics department
at KATH aged <14 years who were being investigated for presumed
pulmonary TB (PTB) based on World Health Organization (WHO)
criteria such as cough, especially if persistent and not resolving,
prolonged fever with or without night sweats, not eating well or
anorexia, weight loss or failure to thrive, unusual fatigue, reduced
playfulness or decreased activity.[10]
A medical history was obtained, and physical examination was
performed. We excluded children with severe hypoxia (oxygen
saturation <92% on supplemental oxygen), severe bronchospasm,
seizures or inability to protect their airways, and those who tested
positive for COVID-19.
Children who had breathing diculty, central cyanosis and/or
wheezing on admission were given respiratory support, and all other
necessary emergency procedures were initiated. They were then
monitored closely until their signs of respiratory distress improved
or resolved, aer which they were assessed for tness to undergo the
IS procedure.
The parents or legal guardians were given information on the
study procedure, and consent was obtained. Assent was also obtained
from children aged >8 years. e nutritional status of the child was
estimated as the weight-for-height z-score using sex, date of birth,
weight and height. In addition, the WHO 2017 classification of
nutritional status of infants and children was used to derive the body
mass index. Weight and height were measured with a Seca scale,
model no. 813, and a stadiometer, model no. 213 (Seca, Germany).
Safety assessment
Sputum induction was performed in a dedicated sputum induction
room aer the patient had fasted for 4 hours. Oxygen saturation and
pulse rate were monitored with a Rad 4 pulse oximeter (Masimo, USA)
before and 30 minutes to 1 hour aer the procedure. e respiratory
rate was also measured over a minute before and aer the procedure
and documented.
Children were nebulised with 2.5 - 5 mg salbutamol and 1 - 2 mL
sterile 3% hypertonic saline attached to a nebuliser for 5 minutes.
Sputum was then suctioned through the nasopharynx with a sterile
mucus extractor with a catheter, size 6 or 8. Older children who were
able to expectorate were encouraged to do so aer nebulisation. A
Table 1. Demographic and anthropometric characteristics of
the study participants (N=144)
Characteristic n (%)*
Age (years)
Median (IQR) 2.5 (0.9 - 6.8)
<2 71 (49.3)
2 - 5 35 (24.3)
6 - 10 27 (18.8)
>10 11 (7.6)
Gender
Female 59 (41.0)
BMI
0 - 5 years (n=98)
Severe acute malnutrition 34 (34.7)
Moderate acute malnutrition 15 (15.3)
Normal 44 (44.3)
Overweight 2 (2.0)
Obese 3 (3.1)
6 - 19 years (n=40)
Severe thinness 15 (37.5)
inness 5 (12.5)
Normal 15 (37.5)
Overweight 2 (5)
Obesity 3 (7.5)
IQR = interquartile range; BMI = body mass index.
*Except where otherwise indicated.
154 AJTCCM VOL. 30 NO. 4 2024
ORIGINAL RESEARCH: ARTICLES
second IS specimen was taken within 48 hours aer the rst sample
had been obtained. No second sample was obtained if the child had
been discharged, had died or was noted to be critically ill. Nebulising
masks were used once for each child and disposed of after the
procedure. Suction tubing was also disposed of aer a single use. e
procedure was terminated if the child’s oxygen saturation dropped
during the procedure.
All sputum specimens were transported immediately to the
Department of Microbiology laboratory at KATH for storage in a 4°C
fridge. Batches of specimens were transported twice weekly in a cooler
box to the Kumasi Centre for Collaborative Research for Xpert MTB/
RIF Ultra and microscopy testing (Ziehl-Neelsen and Auramine O
staining). Mycobacterial cultures were not performed, as culture media
were not available countrywide.
Ethical considerations
Ethical approval was obtained from the Committee on Human Research
Ethics of the School of Medicine and Dentistry, Kwame Nkrumah
University of Science and Technology (ref. no. CHRPE/AP/588/21) and
the Institutional Review Board at KATH (ref. no. KATH/IRB/AP/137/21).
Statistical analysis
Data were entered into a Research Electronic Data Capture (REDCap)
online database and exported to Stata SE 17.0 statistical software
(StataCorp, USA) for analysis aer data cleaning. Summary descriptive
statistics were used for all baseline characteristics. e paired t-test was
used to test the mean dierence in the vital signs before and aer the
IS procedure. e proportions of positive IS samples on GeneXpert
testing, Ziehl-Neelsen staining and Auramine O staining for rst and
second samples were also presented.
Results
From January to June 2022, 144 children, of whom 85 (59.0%) were
male, were enrolled in the study. Nearly half of the participants were
aged <2 years (49.3%), and the median (interquartile range (IQR)) age
was 2.5 (0.9 - 6.8) years (Table 1). Of the 144 participants, 49 (34.0%)
had breathing diculty on admission, and 55 (38.2%) had lower chest
wall retractions. e median (IQR) oxygen saturation on admission was
97% (94 - 99%), the median pulse rate 135 (115 - 154) bpm, and the
median temperature 36.9oC (36.4 - 37.8oC). ree participants (2.1%)
presented with stridor, and 3 (2.1%) with central cyanosis (Table 2).
Sputum induction was successfully performed in all 144 participants
a day or 2 days aer the patient’s condition had normalised or improved,
with approximately two-thirds of the children sampled twice. Epistaxis
was the main adverse eect observed, occurring in 3 participants (2.1%).
It was intermittent and ceased within 30 minutes aer the procedure.
Overall, for all the children sampled there were no clinically signicant
changes observed in temperature, respiratory rate, pulse rate or oxygen
saturation before and aer the procedure. e clinical parameters before
and aer IS are shown in Tables 2 and 3, and the adverse eects in Table 2.
Cumulatively, 98 of the children (68.1%) tested positive for PTB on
Xpert Ultra, comprising 86/144 rst samples (59.7%) and 73/88 second
samples (83.0%) (Table 4). Of the 98 samples that were positive on Xpert
Ultra, 19 (19.4%) were rifampicin resistant (drug resistant).
In addition, 102 of the rst sputum samples and 89 of the second
samples were tested for the presence of MTB using older methods,
Auramine O and Ziehl-Neelsen stains on microscopy. Cumulatively,
Auramine O stains were positive in 57 samples (55.9%) and Ziehl-
Neelsen stains were positive in 47 (46.1%) (Table 4). All the children
who tested positive on Xpert Ultra were successfully initiated on anti-
TB medications.
Discussion
This study has shown that the IS procedure is a safe and effective
diagnostic tool, with minimal adverse eects and well tolerated in
young children, and with a high microbiological yield. Despite the
late introduction of IS in Ghana in 2022 owing to concerns about
patient safety and lack of expertise in performing the procedure, the
Table 2. Clinical parameters on arrival and before and aer the IS procedure (N=144)
Clinical parameter On arrival, n (%)*
Before procedure, n (%)*
Aer procedure, n (%)*
p-value
Fast breathing 49 (34.0) 0 0
Central cyanosis 3 (2.1) 0 0
Chest wall retractions 55 (38.2) 0 0
Pallor 42 (29.2) 0 0
Stridor 3 (2.1) 0 0
Wheezing 27 (18.8) 0 0
Epistaxis 0 0 3 (2.1)
Temperature (oC), median (IQR)
36.9
(36.4 - 37.8)
36.5
(36.5 -37.5)
36.5
(36.2 - 37.1) 0.618
Respiratory rate (cpm), median (IQR)
39
(30 - 50)
38
(30 - 48)
33
(30 - 42) 0.871
Pulse rate (bpm), median (IQR)
135
(115 - 154)
120
(106 - 139)
125
(112 - 142) 0.210
Oxygen saturation in room air (%), median
(IQR)
97 (
90 - 99)
98
(92 - 99)
98
(93 - 99) 0.087
IS = induced sputum; IQR = interquartile range; cpm = cycles per minute.
*Except where otherwise indicated.
AJTCCM VOL. 30 NO. 4 2024 155
ORIGINAL RESEARCH: ARTICLES
study supports reports on IS in many LMICs, such as South Africa,
ailand, Zambia, Mali, e Gambia and Senegal, that found the
procedure to be safe and well tolerated.[6-8,11] We have shown that IS
can be performed in a low-resource setting, as it requires minimal
equipment and expertise to perform with the requisite training.
e IS procedure was successfully performed in all cases, with no
life-threatening complications. Epistaxis was the main adverse eect
observed, and no specialised intervention was required, as all cases
resolved in <30 minutes aer the procedure. Vital signs recorded
before and aer the procedure were similar and within the normal
range, with no statistically significant changes. Our findings are
similar to those of other studies that also reported limited occurrence
of episodes of vomiting, wheezing and epistaxis in a small number of
children undergoing the IS procedure.[7,12,13] e implications of our
ndings and ndings reported by others are that the IS procedure is
a safe and well-tolerated way to obtain samples for testing for TB in
children from whom a sample would otherwise not be available. In
addition, we have shown that the procedure does not require high-
level expertise and can be performed by well-trained nurses. Similar
studies have reported on the feasibility and ease of the procedure
performed by healthcare workers.[8,14]
Furthermore, we have shown that IS can help to confirm the
diagnosis of TB by providing appropriate sputum samples. It can
provide samples to which a newer diagnostic method, GeneXpert, and
older methods such as Auramine O and Ziehl-Neelsen staining using
microscopy, can both be applied.
e most important advantage of the IS procedure is that it can
help increase the number of conrmed diagnoses of PTB, as has
been reported in other centres.[11,12] IS therefore provides an added
advantage to the clinical diagnosis of PTB that has generally been used
in children because of the lack of quality sputum samples for testing,
particularly in young children. Children with presumed TB therefore
have a better chance of a conrmed diagnosis.
As well as supporting microbial conrmation, IS makes it possible
to identify children with drug resistance, as we noted in this study.
When only a clinical diagnosis of TB is made, and there is no sputum
sample, microbial conrmation and conrmation of drug sensitivity
or resistance are difficult to document. An additional long-term
benet of IS is that it helps to promote follow-up of patients with a
positive GeneXpert test with more condence.
e limitation of the present study is that it was carried out in one
centre, and the results therefore cannot be generalised to all healthcare
settings. However, there is potential to expand training in the IS
procedure to other healthcare facilities, and use of the technique in
children with presumed TB across Ghana remains very promising.
Conclusion
e IS procedure is safe and well tolerated in the paediatric population,
and it can increase the yield for microbial conrmation. GeneXpert
testing of IS samples provides information on the sensitivity of MTB
Table 3. Clinical parameters before and aer the IS procedure, according to Xpert Ultra status (N=144)
Xpert Ultra status
Variable, median (IQR) Negative Positive p-value
Examination before IS
Temperature (oC) 36.5 (36.4 - 37.5) 36.6 (36.2 - 37.2) 0.743
Respiratory rate (cpm) 34 (24 - 50) 32 (28 - 42) 0.911
Pulse rate (bpm) 114 (105 - 130) 120 (106 - 140) 0.149
Oxygen saturation (%) 98 (96 - 99) 99 (96 - 99) 0.633
Examination 30 minutes to 1 hour aer IS
Temperature (oC) 36.5 (36.3 - 36.8) 36.5 (36.2 - 37.2) 0.618
Respiratory rate (cpm) 34 (28 - 48) 32 (30 - 42) 0.871
Pulse rate (bpm) 126 (15 - 142) 120 (110 - 140) 0.210
Oxygen saturation (%) 99 (95 - 100) 98 (95 - 99) 0.087
IS = induced sputum; IQR = interquartile range; cpm = cycles per minute.
Table 4. Positivity of GeneXpert MTB/RIF Ultra, Zeihl-
Neelsen staining and Auramine O staining in detecting
Mycobacterium tuberculosis
Test First samples,
n (%)
Second samples,
n (%)
GeneXpert
Positive 86 (60) 73 (83.0)
Negative 58 (40) 15 (17.1)
Total 144 (100) 88 (100)
Ziehl-Neelsen
Positive 38 (37.3) 18 (20.2)
Negative 64 (62.7) 71 (79.8)
Total 102 (100) 89 (100)
Auramine O
Positive 40 (39.2) 34 (38.2)
Negative 62 (60.8) 55 (61.8)
Total 102 (100) 89 (100)
156 AJTCCM VOL. 30 NO. 4 2024
ORIGINAL RESEARCH: ARTICLES
to rifampicin, which is of major importance when a diagnosis of TB
is made in a child in an LMIC setting.
Data availability. e datasets generated and analysed during the present
study are available from the corresponding author (SKO) on reasonable
request.
Declaration. SKO and HJZ are members of the editorial board. The
research for this study was done in partial fulfilment of the
requirements for SKO’s MPhil degree at Kwame Nkrumah University of
Science and Technology, Ghana.
Acknowledgements. The authors wish to acknowledge all the families
that participated in this research.
Author contributions. KSO: project conceptualisation, data
collection, manuscript draft and final review. SKO: project
conceptualisation, data collection, manuscript draft and final
review. NW-B: project conceptualisation, final manuscript review.
EO: data collection. EA: data collection. FAA: data collection. AE:
project conceptualisation, final manuscript review. JS: project
conceptualisation, final manuscript review. OAO: project
conceptualisation, manuscript draft and final review. DMG:
project conceptualisation, final manuscript review. DA: project
conceptualisation, manuscript draft and final review. HJZ: project
conceptualisation, manuscript draft and final review.
Funding.KNUST Research Fund.
Conflicts of interest.None.
1. World Health Organization. Global Tuberculosis Report 2023. Geneva: WHO, 2023.
https://www.who.int/teams/global-tuberculosis-programme/tb-reports/global-
tuberculosis-report-2023 (accessed 17 December 2023).
2. Atherton RR, Cresswell FV, Ellis J, Kitaka SB, Boulware DR. Xpert MTB/RIF Ultra
for tuberculosis testing in children: A mini-review and commentary. Front Pediatr
2019;7:34. https://doi.org/10.3389/fped.2019.00034
3. Kay AW, Ness T, Verkuijl SE, et al. Xpert MTB/RIF Ultra assay for tuberculosis disease
and rifampicin resistance in children. Cochrane Database Syst Rev 2022, Issue 9. Art.
No.: CD013359. https://doi.org/10.1002/14651858.CD013359.pub3
4. DeLuca AN, Hammitt LL, Kim J, et al. Safety of induced sputum collection in children
hospitalised with severe or very severe pneumonia. Clin Infect Dis 2017;64(suppl
3):S301-S308. https://doi.org/10.1093/cid/cix078
5. Zar HJ, Tannenbaum E, Hanslo D, Hussey G. Sputum induction as a diagnostic tool
for community-acquired pneumonia in infants and young children from a high
HIV prevalence area. Pediatr Pulmonol 2003;36(1):58-62. https://doi.org/10.1002/
ppul.10302
6. Ugarte-Gil C, Elkington PT, Gotuzzo E, Friedland JS, Moore DA. Induced sputum is
safe and well-tolerated for TB diagnosis in a resource-poor primary healthcare setting.
Am J Trop Med Hyg 2015;92(3):633-635. https://doi.org/10.4269/ajtmh.14-0583
7. Geldenhuys HD, Kleynhans W, Buckereld N, et al. Safety and tolerability of sputum
induction in adolescents and adults with suspected pulmonary tuberculosis. Eur J Clin
Microbiol Infect Dis 2012;31(4):529-537. https://doi.org/10.1007/s10096-011-1344-5
8. Planting NS, Visser GL, Nicol MP, Workman L, Isaacs W, Zar HJ. Safety and ecacy
of induced sputum in young children hospitalised with suspected pulmonary
tuberculosis. Int J Tuberc Lung Dis 2014;18(1):8-12. https://doi.org/10.5588/
ijtld.13.0132
9. Fahy JV, Boushey HA, Lazarus SC, et al. Safety and reproducibility of sputum
induction in asthmatic subjects in a multicenter study. Am J Respir Crit Care Med
2001;163(6):1470-1475. https://doi.org/10.1164/ajrccm.163.6.9901105
10. World Health Organization. WHO consolidated guidelines on tuberculosis: Module
5: Management of tuberculosis in children and adolescents. 18 March 2022. https://
www.who.int/publications-detail-redirect/9789240046764 (accessed 31 March 2024).
11. Zar HJ, Hanslo D, Apolles P, Swingler G, Hussey G. Induced sputum versus gastric
lavage for microbiological conrmation of pulmonary tuberculosis in infants and
young children: A prospective study. Lancet 2005;365(9454):130-134. https://doi.
org/10.1016/S0140-6736(05)17702-2
12. Moore HA, Apolles P, de Villiers PJT, Zar HJ. Sputum induction for microbiological
diagnosis of childhood pulmonary tuberculosis in a community setting. Int J Tuberc
Lung Dis 2011;15(9):1185-1190. https://doi.org/10.5588/ijtld.10.0681
13. Zar HJ, Tannenbaum E, Apolles P, Roux P, Hanslo D, Hussey G. Sputum induction for
the diagnosis of pulmonary tuberculosis in infants and young children in an urban
setting in South Africa. Arch Dis Child 2000;82(4):305-308. https://doi.org/10.1136/
adc.82.4.305
14. Nicol MP, Zar HJ. New specimens and laboratory diagnostics for childhood
pulmonary TB: Progress and prospects. Paediatr Respir Rev 2011;12(1):16-21. https://
doi.org/10.1016/j.prrv.2010.09.008
Received 14 June 2024. Accepted 20 October 2024. Published 10 December 2024.
