; Sputum induction for the diagnosis of pulmonary tuberculosis in
Learning Center
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Sputum induction for the diagnosis of pulmonary tuberculosis in


  • pg 1
									                          Downloaded from adc.bmj.com on December 7, 2011 - Published by group.bmj.com

Arch Dis Child 2000;82:305–308                                                                                                305

                             Sputum induction for the diagnosis of pulmonary
                             tuberculosis in infants and young children in an
                             urban setting in South Africa
                             H J Zar, E Tannenbaum, P Apolles, P Roux, D Hanslo, G Hussey

                             Abstract                                          there are no reports of its use in infants or chil-
                             Background—Bacteriological confirma-               dren younger than 3 years.
                             tion of pulmonary tuberculosis is diYcult            Gastric lavage (GL) is regarded as the stand-
                             in infants and young children. In adults          ard procedure to obtain specimens for staining
                             and older children, sputum induction has          and culture of Mycobacterium tuberculosis in
                             been successfully used; this technique has        younger children because they swallow their
                             not been tested in younger children.              sputum and do not expectorate. Studies
                             Aims—To investigate whether sputum in-            comparing GL to induced sputum in adults
                             duction can be successfully performed in          with suspected tuberculosis reported induced
                             infants and young children and to deter-          sputum to be more eVective4 5; however, one
                             mine the utility of induced sputum com-           study reported that if GL was undertaken after
                             pared to gastric lavage (GL) for the              sputum induction the procedures produced
                             diagnosis of pulmonary tuberculosis in            similar results.6 A study of 13 children with
                             HIV infected and uninfected children.             pulmonary tuberculosis reported an improved
                             Subjects and methods—149 children (me-            yield from GL when preceded by a nebulisa-
                             dian age 9 months) admitted to hospital           tion of superheated isotonic saline.7 There are
                             with acute pneumonia who were known to            no published studies comparing induced spu-
                             be HIV infected, suspected to have HIV            tum to GL for culture of M tuberculosis in chil-
                             infection, or required intensive care unit        dren. Two studies of GL compared to broncho-
                             support. Sputum induction was per-                alveolar lavage (BAL) in children suggest that
                             formed on enrolment. Early morning GL             GL provides a higher bacteriological yield than
                             was performed after a minimum four hour           BAL for pulmonary tuberculosis.8 9
                             fast. Induced sputum and stomach con-                The aim of this study was to determine
                             tents were stained for acid fast bacilli and      whether sputum induction can be successfully
                             cultured for Mycobacterium tuberculosis.          performed in infants and young children, to
                             Results—Sputum induction was success-             compare induced sputum with GL for the yield
                             fully performed in 142 of 149 children. M         of M tuberculosis in children with pulmonary
                             tuberculosis, cultured in 16 children, grew       tuberculosis, and to determine whether the
                             from induced sputum in 15. GL, per-               yield was influenced by HIV status.
                             formed in 142 children, was positive in
                             nine; in eight of these M tuberculosis also
                             grew from induced sputum. The diVer-              Methods
                             ence between yields from induced sputum
                             compared to GL was 4.3% (p = 0.08). M             A prospective one year study during 1998 was
                             tuberculosis was cultured in 10 of 100 HIV        performed in the paediatric wards of four hos-
                             infected children compared to six of 42           pitals in Cape Town, South Africa—Red Cross
                             HIV uninfected children (p = 0.46).               War Memorial Children’s Hospital, Somerset,
                             Conclusion—Sputum induction can be                Conradie, and Groote Schuur. Children en-
                             safely and eVectively performed in infants        rolled in this study were participants in a larger
Department of
                             and young children. Induced sputum pro-           study to determine the aetiology of acute
Paediatrics and Child                                                          pneumonia in hospitalised, HIV infected chil-
Health, Red Cross War        vides a satisfactory and more convenient
                                                                               dren. Children with a primary diagnosis of
Memorial Children’s          specimen for bacteriological confirmation
                                                                               pneumonia according to World Health Organ-
Hospital, University of      of pulmonary tuberculosis in HIV infected
Cape Town, South                                                               isation criteria10 (defined as the presence of
                             and uninfected children.
Africa                       (Arch Dis Child 2000;82:305–308)                  tachypnoea or lower chest indrawing) and who
H J Zar                                                                        were known to be HIV infected, were sus-
E Tannenbaum                 Keywords: induced sputum; tuberculosis; HIV       pected of having HIV infection, or were admit-
P Apolles                                                                      ted to the intensive care unit (ICU) but were
P Roux
                                                                               not intubated were studied. A suspicion of HIV
D Hanslo
G Hussey                     Bacteriological confirmation of pulmonary          infection was based on the presence (in
                             tuberculosis in infants and children remains      addition to pneumonia) of two or more of the
Correspondence to:           diYcult. Sputum induction has been used to        following:     generalised    lymphadenopathy,
Dr H Zar, Child Health       diagnose pulmonary tuberculosis in HIV in-        weight below the 3rd centile for age, hepatome-
Unit, 46 Sawkins Road,
Rondebosch, 7700, South      fected and immunocompetent adults.1 Older         galy, splenomegaly, oral candidiasis, enlarged
Africa                       children can produce or be induced to produce     parotid glands, or chronic diarrhoea. Children
email: heather@              sputum; this method has been used for the         were enrolled during working hours from
                             diagnosis of Pneumocystis carinii pneumonia       Monday to Friday. Informed consent for
Accepted 24 November 1999    and more recently for tuberculosis.2 3 However,   enrolment in the study and for HIV testing (in
      Downloaded from adc.bmj.com on December 7, 2011 - Published by group.bmj.com

306                                                                                  Zar, Tannenbaum, Apolles, Roux, Hanslo, Hussey

         Table 1 Characteristics of children in whom sputum                  oropharynx using a sterile, mucus extractor of
         induction was performed                                             catheter size 6. Specimens were transported
                                          Children without Children with
                                                                             directly to the laboratory for processing.
         Characteristic                   TB (n =126)      TB (n = 16)
                                                                             LABORATORY METHODS
         Age (months)                     9 (3–21.5)        12 (7–25)
         Male:female                      1.2               2.2              GL specimens collected on consecutive days
         ICU admission, n (%)             18 (14)           2 (12)           were pooled before culturing, while induced
         HIV positive, n (%)              90 (71)           10 (63)
         TB contact, n (%)                24 (19)           7 (44)*
                                                                             sputum samples were cultured singly. Speci-
         Use of supplemental O2, n (%)    83 (66)           8 (50)           mens were decontaminated with sodium hy-
         Baseline O2 saturation in air    94 (90.5–97)      93.5 (88–98)     droxide, and after neutralisation with buVered
         Baseline RR                      50 (40–60)        53 (40–63.5)
                                                                             saline were concentrated by centrifugation.
         Continuous variables expressed as median (25th to 75th              The resuspended deposit was used to make a
         percentile).                                                        smear for microscopy. Prepared smears from
         TB, tuberculosis; ICU, intensive care unit; RR, respiratory rate.
         *p = 0.02.
                                                                             the concentrated samples were stained with
                                                                             Auramine-O and examined by fluorescence
         children with suspected HIV infection or                            microscopy for the presence of acid fast bacilli.
         admitted to ICU) was obtained from a parent.                        A Bactec 12B bottle (Becton Dickinson, USA)
         The study was approved by the Research and                          containing supplemented Middlebrook me-
         Ethics Committee of the University of Cape                          dium was inoculated with 0.5 ml of sample,
         Town.                                                               incubated at 37°C, and monitored for growth
            A history and physical examination were                          twice weekly for the first two weeks, then
         performed in every child enrolled. HIV infec-                       weekly for a total incubation period of six
         tion was confirmed by two positive enzyme                            weeks. Positive mycobacterial growth was con-
         linked immunosorbent assay (ELISA) tests                            firmed by a stain for acid fast bacilli as well as
         (Vironostika HIV Uniform II, Organon Te-                            by PCR using primers prepared in house.
         knika, Holland) in children older than 18
         months or by a positive ELISA and polymerase                        STATISTICAL ANALYSIS
         chain reaction (PCR; Amplicor HIV-1, Roche                          Results were analysed using Epi6.04 (CDC,
         Diagnostic Systems) in younger children.                            Atlanta, USA). Comparison between HIV
         Tuberculosis was diagnosed only when cul-                           infected and uninfected children was made
         tures of induced sputum or GL grew M tuber-                         using the 2 and Kruskal-Wallis tests where
         culosis.                                                            appropriate. The yield of M tuberculosis from
                                                                             induced sputum and gastric lavage was com-
         GASTRIC LAVAGE                                                      pared using the McNemar test.
         Early morning GL was performed in children
         younger than 5 years old after an overnight fast                    Results
         of at least four hours. A nasogastric tube was                      Sputum induction was successfully performed
         passed before the child arose and the gastric                       in 142 of 149 children with a median (25th to
         contents aspirated. Normal saline 20 ml was                         75th centile) age of 9 (3–20) months (table 1).
         inserted down the tube, left for two to three                       The youngest child in whom sputum was suc-
         minutes, and then aspirated. Additional 5–10                        cessfully obtained was 1 month old. Seven
         ml normal saline aliquots were inserted and                         children were either considered too ill to toler-
         aspirated until a minimum of 20 ml of aspirate                      ate sputum induction or developed increasing
         was obtained. Gastric aspirates were placed in a                    tachypnoea or cough during nebulisation or
         sterile, sodium carbonate containing tube and                       suctioning, necessitating termination of the
         taken to the laboratory. GL was ideally                             procedure. Only a minority of children (10%)
         performed on two or three consecutive morn-                         could expectorate and sputum was obtained by
         ings; however, repeated lavages were not possi-                     suctioning in the remainder. No serious
         ble in many children owing to other factors                         adverse reactions occurred during sputum
         such as discharge from hospital or subsequent                       induction but there were minor events includ-
         intubation.                                                         ing mild epistaxis in six, increase in coughing in
                                                                             eight, and wheezing responsive to an inhaled
         SPUTUM INDUCTION                                                    bronchodilator in three. The baseline median
         Sputum induction was undertaken on the day                          respiratory rate of children was 50 (40–60) and
         of enrolment after a two to three hour fast. A                      arterial oxygen saturation in air was 94% (92–
         physiotherapist or research nurse trained in the                    97%). One hundred children were found to be
         use of this technique performed the procedure.                      HIV infected.
         Children were pretreated with 200 µg salbuta-                          M tuberculosis was cultured from sputum or
         mol given via metered dose inhaler with                             GL in 16 children. Sputum cultures grew M
         attached spacer (Babyhaler, GlaxoWellcome)                          tuberculosis in 15 of 142 children but GL was
         to prevent the occurrence of broncho-                               positive in only nine (table 2). A single GL
         constriction.11 A jet nebuliser (GRS, Inter-                        sample was obtained in 39 children (of which
         surgical, UK) attached to oxygen at a flow rate                      two were positive); there were two pooled
         of 5 l/min delivered 5 ml of 5% sterile saline for                  lavage specimens in 77 (four positive), and
         15 minutes. Thereafter physiotherapy tech-                          three pooled specimens in 26 (three positive).
         niques including chest percussion, vibration,                       In only one case was the GL culture positive for
         shaking, and active cycle breathing were                            mycobacteria while the corresponding sputum
         applied. Sputum was obtained either by expec-                       was negative. The diVerence (95% confidence
         toration (in children able to cooperate) or by                      interval) between yields for M tuberculosis from
         suctioning through the nasopharynx or                               culture of induced sputum compared to GL
                          Downloaded from adc.bmj.com on December 7, 2011 - Published by group.bmj.com

Diagnosis of pulmonary tuberculosis in South Africa                                                                                      307

                              Table 2 Diagnosis of children with tuberculosis by sputum     and importance of the skin tuberculin test. In
                              induction or gastric lavage                                   addition, other HIV associated lung diseases
                                                      Gastric lavage
                                                                                            may mimic the clinical and radiological picture
                                                                                            of pulmonary tuberculosis. Gastric aspirates
                                                      Culture positive   Culture negative   have been reported to give isolation rates of M
                              Induced sputa           (n)                (n)
                                                                                            tuberculosis ranging from 28% to 40% in
                              Culture positive (n)    8                    7                children with suspected tuberculosis, although
                              Culture negative (n)    1                  126                higher rates of up to 75% have been described
                                                                                            in infants.12–14 Although GL was performed
                              specimen was 4.3% (0% to 5.6%; p = 0.08).                     according to a standard protocol, diVerent
                              Of the specimens culture positive for M                       nurses performed this procedure as an early
                              tuberculosis, three of 15 induced sputum (20%)                morning preprandial specimen was required.
                              and three of nine GL specimens (33%) were                     The relatively lower culture rate of M tuberculo-
                              positive by microscopy for acid fast bacilli.                 sis from GL compared to sputum induction
                                The median age of children with bacterio-                   may therefore reflect some variability in GL
                              logical confirmation of tuberculosis was 12                    technique and the lack of well standardised
                              (7–25) months which was similar to those                      guidelines for the procedure.
                              without tuberculosis (9 (3–21.5) months;                         Sputum induction mobilises secretions from
                              p = 0.19). The youngest child in whom M                       the lower respiratory tract. Hypertonic saline,
                              tuberculosis was cultured from induced sputum                 deposited in the lower airways, causes intersti-
                              was 3 months of age while seven of 15 children                tial fluid to move into this area by osmosis.15
                              were less than 1 year. No child had cavitatory                Furthermore, hypertonic saline stimulates the
                              tuberculosis. Ten children culture positive for               cough reflex, causing secretions from the lower
                              M tuberculosis were HIV infected. This repre-                 respiratory tract to be moved upwards. The
                              sents 10% of all HIV infected children, which is              eYcacy of sputum induction in this study may
                              similar to the percentage of HIV negative chil-               be a result of the application of a standard
                              dren with tuberculosis (six of 42, 14.3%) (odds               technique by a few people trained in its use,
                              ratio 0.67 (0.2–2.41), p = 0.46).                             and to rapid processing of the specimen by the
                                                                                            laboratory. Nevertheless, this technique is rela-
                                                                                            tively simple, does not require sophisticated
                              Discussion                                                    equipment, and can be taught to health care
                              This study showed that sputum induction can                   workers. Care should be taken to ensure that
                              be eVectively and safely performed in infants                 sputum induction does not facilitate the spread
                              and young children including those with HIV                   of tuberculosis to other patients or staV.
                              infection. The procedure could be successfully                Sputum induction in children is unlikely to
                              performed in infants as young as 1 month of                   result in tuberculosis transmission owing to the
                              age. Sputum induction was well tolerated even                 paucibacillary nature of their disease as evi-
                              in children who were hypoxic or who had                       denced by a minority of children who were
                              AIDS. Although we were unable to perform                      smear positive. Nevertheless, it is recom-
                              continuous monitoring of arterial oxygen satu-                mended that sputum induction be performed
                              ration during sputum induction, very few chil-                in a room with negative pressure ventilation16;
                              dren were unable to complete the procedure                    in the absence of such a facility (as occurs in
                              and no child subsequently appeared to deterio-                most developing countries), the procedure
                              rate clinically. Sputum induction was a more                  should be done in a well ventilated room and
                              sensitive method than GL for culture of M                     equipment sterilised between patients.
                              tuberculosis, detecting almost twice the number               Whereas GL is time consuming, distressing to
                              of children with pulmonary tuberculosis. The                  the child and care giver, and should be
                              lack of statistically significant diVerence be-                repeated on consecutive days, induced sputum
                              tween the yield from induced sputum com-                      is easier to perform, relatively non-invasive,
                              pared to GL is most likely a result of the small              does not require hospitalisation, and can be
                              number of children with tuberculosis and a                    repeated if necessary. Although GL can be suc-
                              larger study would be useful. The small                       cessfully performed in an ambulatory setting,17
                              number of tuberculosis cases reflects the                      the majority of children require hospitalisation;
                              relatively low risk of infection among the study              in contrast sputum induction is easily done as
                              group as they were hospitalised for acute pneu-               an outpatient procedure.
                              monia. Nevertheless, pulmonary tuberculosis                      In conclusion, sputum induction can be
                              was a common diagnosis in this group of chil-                 eVectively performed and is well tolerated and
                              dren from a high HIV prevalence population                    safe even in infants. Induced sputum is better
                              who had clinical features of acute respiratory                than GL for the isolation of M tuberculosis in
                              infection. Further study of the eYcacy of spu-                both HIV infected and uninfected infants and
                              tum induction in children presenting with                     children. The bacteriological yield from spu-
                              chronic illness suggestive of tuberculosis, such              tum or GL for pulmonary tuberculosis does
                              as failure to thrive and persistent fever, is war-            not diVer by HIV status. Use of induced
                              ranted.                                                       sputum should be considered as a first line
                                 Childhood tuberculosis is an increasingly                  investigation in children suspected of having
                              important public health problem, particularly                 pulmonary tuberculosis, especially in circum-
                              in developing countries. The diagnosis of                     stances in which a culture confirmed diagnosis
                              tuberculosis is notoriously diYcult in children,              should be vigorously sought (such as when the
                              especially those who are HIV infected, owing to               source case is unknown, drug resistance is sus-
                              the development of anergy which limits the use                pected, or cutaneous anergy occurs).
      Downloaded from adc.bmj.com on December 7, 2011 - Published by group.bmj.com

308                                                                                   Zar, Tannenbaum, Apolles, Roux, Hanslo, Hussey

         This work was funded by the MRC, South Africa, an ASTRA             8 Abadco DL, Steiner P. Gastric lavage is better than
         Respiratory Fellowship (HZ) awarded by the South African               bronchoalveolar lavage for isolation of Mycobacterium
         Pulmonology Society, and the ICH Fund, Red Cross Children’s            tuberculosis in childhood pulmonary tuberculosis. Pediatr
         Hospital. We thank the laboratory staV, interpreters, pharma-          Infect Dis J 1992;11:735–8.
         cists, social workers, ward doctors, nursing staV, and Ms W         9 Somu N, Swaminathan S, Paramasivan CN, et al. Value of
         Isaacs for help with administering the study. We acknowledge           bronchoalveolar lavage and gastric lavage in the diagnosis
         Dr George Swingler for advice on statistics and review of the          of pulmonary tuberculosis in children. Tuber Lung Dis
         manuscript. Finally, we thank the children and their parents for       1995;76:295–9.
         participating in the study.                                        10 World Health Organisation Programme for the Control of
                                                                                Acute Respiratory Infections. ARI programme management.
          1 Parry CM, Kamoto O, Harries AD, et al. The use of sputum            A training course. Module on National Policies. Geneva: World
             induction for establishing a diagnosis in patients with sus-       Health Organisation, 1990.
             pected pulmonary tuberculosis in Malawi. Tuber Lung Dis        11 Belcher NG, Lee TH, Rees PJ. Airway responses to
             1995;76:72–6.                                                      hypertonic saline, exercise and histamine challenges in
          2 Ognibene FP, Gill VJ, Pizzo PA, et al. Induced sputum to            bronchial asthma. Eur Respir J 1989;2:44–8.
             diagnose Pneumocystis carinii pneumonia in immunosup-          12 Lloyd AVC. Bacteriological diagnosis of tuberculosis in
             pressed pediatric patients. J Pediatr 1989;115:430–3.              children. A comparative study of gastric lavage and
          3 Shata AMA, Coulter JBS, Parry CM, et al. Sputum                     laryngeal swab methods. East Afr Med J 1968;45:140–3.
             induction for the diagnosis of tuberculosis. Arch Dis Child    13 Starke JR, Yaylor-Watts KT. Tuberculosis in the pediatric
             1996;74:535–7.                                                     population of Houston, Texas. Pediatrics 1989;84:28–35.
          4 Jones FL. The relative eYcacy of spontaneous sputa,             14 Vallejo JG, Ong LT, Starke JR. Clinical features, diagnosis
             aerosol-induced sputa and gastric aspirates in the bacterio-       and treatment of tuberculosis in infants. Pediatrics 1994;84:
             logic diagnosis of pulmonary tuberculosis. Dis Chest 1966;         1–7.
          5 Lillehei JP. Sputum induction with heated aerosol inhala-       15 O’Byrne P, Hargreave F. Non-invasive monitoring of airway
             tions for the diagnosis of tuberculosis. Am Rev Respir Dis         inflammation. Am J Respir Crit Care Med 1994;150:S100-2.
             1961;84:276–8.                                                 16 Centres for Disease Control. Guidelines for preventing the
          6 Carr DT, Karlson AG, Stilwell GG. A comparison of                   transmission of tuberculosis in health-care settings with
             cultures of induced sputum and gastric washings in the             special focus on HIV-related issues. MMWR Morb Mortal
             diagnosis of tuberculosis. Mayo Clin Proc 1967;42:23–5.            Wkly Rep 1990;39:1–29.
          7 Giammona ST, Zelkowitz PS. Superheated nebulized saline         17 Lobato MN, LoeZer AM, Furst K, Cole B, Hopewell PC.
             and gastric lavage to obtain bacterial cultures in primary         Detection of Mycobacterium tuberculosis in gastric
             pulmonary tuberculosis in children. Am J Dis Child                 aspirates collected from children: hospitalization is not
             1969;117:198–200.                                                  necessary. Pediatrics 1998:102:e40.
                     Downloaded from adc.bmj.com on December 7, 2011 - Published by group.bmj.com

                                  Sputum induction for the diagnosis of
                                  pulmonary tuberculosis in infants and young
                                  children in an urban setting in South Africa
                                  H J Zar, E Tannenbaum, P Apolles, et al.

                                  Arch Dis Child 2000 82: 305-308
                                  doi: 10.1136/adc.82.4.305

                                  Updated information and services can be found at:

                                  These include:
         References               This article cites 13 articles, 6 of which can be accessed free at:

                                  Article cited in:

     Email alerting               Receive free email alerts when new articles cite this article. Sign up in the
           service                box at the top right corner of the online article.

              Topic               Articles on similar topics can be found in the following collections
                                     Child health (1698 articles)
                                     HIV/AIDS (77 articles)
                                     Immunology (including allergy) (972 articles)
                                     Sexual health (180 articles)
                                     TB and other respiratory infections (310 articles)
                                     Tuberculosis (15 articles)
                                     Infant health (304 articles)
                                     Pneumonia (infectious disease) (94 articles)
                                     Pneumonia (respiratory medicine) (83 articles)
                                     Stomach and duodenum (26 articles)


To request permissions go to:

To order reprints go to:

To subscribe to BMJ go to:

To top