|Year : 2017 | Volume
| Issue : 2 | Page : 106-113
Diagnosing pulmonary tuberculosis in children
Valsan Philip Verghese
Pediatric Infectious Diseases, Department of Paediatrics, Christian Medical College, Vellore, Tamil Nadu, India
|Date of Web Publication||18-May-2017|
Valsan Philip Verghese
Department of Child Health, Christian Medical College, Vellore - 632 004, Tamil Nadu
Source of Support: None, Conflict of Interest: None
Children account for about 5% of new TB cases in India but this may be an underestimate, as clinical signs of TB are subtle in children. Pulmonary TB is the most common form in children but diagnosis is challenging due to various factors including difficulty in collecting specimens from young children who tend to swallow sputum and produce fewer TB bacilli in specimens compared to adults. Tests such as the AFB smear used in adults are therefore often negative in children. A high index of suspicion needs to be maintained in order to make the diagnosis. Symptoms such as persistent fever and non-remitting cough are highly specific for a diagnosis of TB. The chest radiograph is the most important investigation for making an early diagnosis of in pediatric pulmonary TB. When starting empiric anti-tuberculous therapy in children, it is essential to monitor the treatment response in the form of resolution of symptoms and gain in weight. A poor clinical response to treatment should alert one to the possibility of poor adherence to therapy, incorrect doses of antituberculous drugs, presence of TB resistant to first-line drugs or an incorrect diagnosis of tuberculosis in the first place.
Keywords: Tuberculosis in children, sputum AFB, chest X-ray in tuberculosis
|How to cite this article:|
Verghese VP. Diagnosing pulmonary tuberculosis in children. Curr Med Issues 2017;15:106-13
| Introduction|| |
India accounts for nearly a quarter of the world's burden of tuberculosis (TB), with 2.2 million out of an estimated global incidence of 9.6 million newly diagnosed cases of TB every year. While TB prevalence has shown a decline from 465 in 1990 to 195 per 100,000 population in 2013, these figures in a large population such as ours mean that there are close to 25 lakh people living with TB in our country. Children are estimated to account for 5% of new TB cases, but this may be an underestimate as clinical signs of TB are subtle in children. In addition, tests such as the acid-fast bacilli (AFB) smear used in adults are often negative in children due to the difficulty in collecting specimens (as young children tend to swallow sputum) as well as the fact that children produce less TB bacilli compared to adults. This means that a high index of suspicion needs to be maintained to make the diagnosis of TB in children. As most TB in children is pulmonary, this article will focus on the approach to make a diagnosis of pulmonary TB in children (common manifestations of extrapulmonary TB in children are given in Appendix 1).
A 4-year-old girl child is brought to the outpatient clinic with fever for 3. days and no other symptoms. On examination, her weight is 13 kg. (81% expected) and her height is 96 cm. She has some small nonspecific cervical lymph nodes.
Respiratory system examination reveals respiratory rate to be 28/min, no use of accessory muscles of respiration, breath sounds are heard equally in all areas, and no adventitious sounds are heard. Examination of the throat and ear is normal. On abdominal examination, there is no hepatosplenomegaly. Systemic examination is normal.
You diagnose her to have viral fever, give her antipyretics, and ask her father to bring her back in 48 h if fever persists. She is brought back 11. days later still having daily, intermittent, moderate-grade fever.
On examination, 14. days after the onset of fever, her weight is 12.5 kg, respiratory rate is 32/min, and she has no new findings.
Do you think this child has TB?
| Recommended Approach to Diagnose Pulmonary Tuberculosis in Children (Adapted from Childhood Tuberculosis Training Toolkit, World Health Organization 2014)|| |
The approach to make a diagnosis of pulmonary TB  in children is summarized in six steps as follows:
- Symptoms suggestive of TB: does the child have any of these?
- History of TB contact: does anyone in contact with the child have TB or suspected TB?
- Clinical examination (including weight assessment): Are there any clinical signs of TB?
- Tuberculin skin testing (TST/Mantoux): Is it positive or negative?
- Chest X-ray: Are there any chest X-ray findings that could be due to TB?
- Bacteriological confirmation whenever possible: Have these tests been done before starting antituberculous therapy (ATT)?
Step 1: Ask for symptoms suggestive of pulmonary tuberculosis
The three main symptoms that should make a physician suspect pulmonary TB in children are the following:
- Persistent* fever
- Persistent* cough, and
- Weight loss or failure to gain weight.
(*Persistent being defined as lasting for more than 2 weeks without improvement with other therapy, such as broad-spectrum antibiotics).
While an acute cough is suggestive of a viral respiratory infection and recurrent cough suggests asthma, a persistent nonremitting cough is highly specific for a diagnosis of TB, as is persistent fever.
Other symptoms that may indicate TB in children include fatigue and reduced playfulness in a child, but these are less commonly elicited compared to the first three.
Persistent fever or non-remitting cough is highly specific for a diagnosis of TB.
Step 2: Ask for a history of contact with tuberculosis
If there is a history of the child having been in contact with someone with TB, ask for the following:
- How close is the person with TB with whom the child has been in contact with?
Data show that a close (usually household) contact is more likely to transmit TB than a casual (nonhousehold) contact.
- When did the contact occur?
Most children who manifest with signs of TB do so within the first 2 years of exposure to TB, and 90% tend to do so within the 1st year of contact.
- Is the sputum status of the person with TB known?
People with sputum-positive pulmonary TB are more likely than sputum-negative patients to infect children whom they come in contact with. Besides, if a sputum AFB culture susceptibility is available, that would help guide both preventive as well as treatment for the child, while awaiting results of the child's own AFB culture.
Ask the leading question: If no source case with TB is identified, always ask whether anyone in the household has persistent cough or fever. Identifying a household contact is important to protect other children in the house from developing new TB infections as well as preventing reinfection of a child who has completed treatment.
Step 3: Look for signs of pulmonary tuberculosis
The most common clinical finding in primary pulmonary TB in children is a normal respiratory system examination!
However, careful examination including a count of the respiratory rate can reveal the following findings that are associated with certain forms of pulmonary TB:
- Tachypnea in miliary TB
- Dullness and diminished air entry with TB pleural effusion
- Clubbing and coarse crepitation with bronchiectasis secondary to TB
- Asymmetric wheeze with airway compression due to enlarged intrathoracic TB lymph nodes.
Often, however, the only finding is weight loss or failure to gain weight over the symptomatic period.
Step 4: Do a tuberculin skin test (Mantoux)
Although a positive Mantoux test only indicates exposure to TB (called TB infection rather than TB disease), it is added evidence that the symptoms may be due to TB. In the child with a positive Mantoux test, the absence of any evidence of TB disease after a full workup should also alert one to the need to provide prophylaxis with 6 months of isoniazid to prevent developing actual TB disease.
Technique of TST administration:
- 0.1 ml (5 TU) of purified protein derivative in tuberculin syringe with 26-gauge needle injected into the intradermal plane 10 cm below elbow on volar aspect of the left (nondominant) forearm.
- Raise a wheal of 6–10 mm. This usually disappears in 10–15 min.
- Read by measuring the degree of induration (not the erythema) in the transverse plane at 48–72 h.
- A TST reading ≥10 mm is considered positive and a TST ≥5 mm is considered positive in HIV infection or immunosuppression.
Points to note in the interpretation of tuberculin skin testing
TST has poor sensitivity and specificity, with both false-negative tests in malnourished children with TB and false-positive tests in a Bacillus Calmette–Guérin-vaccinated population. This means that a positive TST is a useful added evidence for TB disease ONLY when the child has other findings suggestive of TB. If the child has no other findings of TB such as a history suggestive of TB or clinical or laboratory (radiologic or bacteriologic) evidence of TB, a positive TST can only be taken as evidence of exposure or latent TB infection necessitating prophylaxis with 6 months of isoniazid.
The reverse also holds true that a negative Mantoux/TST does not rule out TB.
Key points - Chest X-ray in pediatric pulmonary TB
- The chest radiograph is the most important investigation for making an early diagnosis of pediatric pulmonary TB
- Hilar or paratracheal lymphadenopathy is the hallmark of primary TB in children
- Chest X-ray findings in childhood TB may overlap with those seen in other disease conditions
- However, X-ray findings that persist despite appropriate therapy. (including antibiotics) should raise a strong suspicion of TB, especially when the child continues to have symptoms suggestive of TB.
A positive tuberculin skin test is a useful added evidence for TB disease ONLY when the child has other findings suggestive of TB.
Step 5: Look at the chest radiograph
The chest radiograph is the most important investigation for making an early diagnosis of pediatric pulmonary TB.
Understanding the various radiological manifestations of TB involves understanding the pathogenesis of primary TB as shown in [Figure 1].
|Figure 1: Progression of pathology of intrathoracic tuberculosis after primary infection.|
Click here to view
Primary pulmonary infection with TB gives rise to the primary complex, comprising the Ghon focus which is single, subpleural, and often seen in the right upper lobe, and the lymph nodes drained by the focus, the paratracheal and perihilar lymph nodes [Figure 2]. As formation of the Ghon complex is often subclinical, the most common radiographic abnormality seen in primary pulmonary TB in young children is isolated hilar adenopathy.
|Figure 2: (a) Primary complex: Ghon focus and regional nodes. (b) Chest X-ray showing Ghon focus (bottom arrow) and hilar lymph node (top arrow).|
Click here to view
In relatively immune-competent children, disease containment can lead to both the focus and nodes being quiescent for years, during which time the focus and node may disappear or calcify. However, as these contain dormant bacilli, reactivation of TB can occur under conditions of stress such as malnutrition or decreased immunity, often seen following viral infections such as measles. In infants and immunocompromised children, poor disease containment can give rise to progressive caseation. A caseous focus can rupture into a bronchus leading to cavity formation [Figure 3] or into the pleural cavity leading to pleural effusion [Figure 4]. A pleural effusion that is often picked up as an incidental finding in older children with prolonged fever may be the only clue that the cause of the fever is TB.
|Figure 3: (a) Rupture of focus into bronchus causing cavitation. (b) Chest X-ray showing rupture of focus into bronchus causing cavitation (bottom arrow) and bilateral bronchopneumonia (top arrow).|
Click here to view
|Figure 4: (a) Rupture of focus into pleural cavity causing effusion. (b) Chest X-ray showing tuberculosis pleural effusion (arrow).|
Click here to view
A pleural effusion that is often picked up as an incidental finding in older children with prolonged fever may be the only clue that the cause of the fever is TB.
Rupture of a cavity or caseous node into a bronchus can lead to intrabronchial spread of infection causing bronchopneumonia [Figure 5]. Hematogenous spread after primary infection in the very young or immunocompromised can lead to the classical picture of miliary TB, with fine uniform infiltrates involving all the lung fields
|Figure 5: (a) Intrabronchial spread from focus or node: Bronchopneumonia.(b) Chest X-ray showing erosion of hilar lymph node (arrow) into left bronchus with left-sided pneumonia.|
Click here to view
Lymphobronchial TB (also called intrathoracic lymph node TB), due to the effect of enlarging lymph nodes on smaller airways of younger children, is most commonly seen following primary infection before 5 years of age. Partial or check-valve obstruction of an airway by a node can lead to hyperinflation while complete obstruction leads to collapse of a segment or lobe distal to the bronchus. Bronchiectasis may be seen in collapsed lobes where the obstruction is irreversible despite treatment [Figure 6],[Figure 7],[Figure 8].
|Figure 6: (a) Partial obstruction of bronchus by node leading to hyperinflation of distal lobes. (b) Chest X-ray showing incomplete obstruction of left main bronchus with hyperinflation of the left lung.|
Click here to view
|Figure 7: (a) Complete obstruction of bronchus by node leading to collapse. (b) Chest X-ray showing lymph node obstructing bronchus with collapse of the right lower lobe (arrow).|
Click here to view
Although the radiologic findings in TB may overlap with those seen in other disease conditions, persistence of X-ray findings despite what appears to be appropriate therapy, including antibiotics, should raise a strong suspicion of TB, especially when the child continues to be symptomatic.
Case Scenario (Continued)
Due to this child's prolonged fever, the physician asks the father whether anyone in the family has any illness suggestive of TB. He says that he has been having a cough for the past 2 months that does not seem to get better despite medicines from his local pharmacy and that he feels feverish at night.
The following tests are done for the child:
- Mantoux test: 2 mm × 3 mm, negative
- Gastric aspirate for AFB testing: AFB smear is negative. AFB culture has been sent and will be ready sometime in the next 3 months.
- Chest X-ray shows a rounded shadow in the right hilar region and paracardiac infiltrates in both lung fields [Figure 9].
Will you start ATT in this child?
Step 6: Send tests for bacterial confirmation
With 2%–3% of new TB cases and 15% of retreated TB cases in India estimated to be multidrug resistant, it is always advisable to send a sample (sputum in older children and gastric aspirates in children below the age of 10 years who cannot expectorate sputum) for mycobacterial testing before giving ATT that often needs to be started before results of confirmatory testing become available.
As TB in children is predominantly paucibacillary, AFB smears on sputum specimens are often poor tests to detect TB, with smear positivity ranging from 38% in detecting clinical TB to 41% in culture-confirmed TB., For the same reason, AFB cultures, the “gold standard” for TB diagnosis, are also positive only in about 30%–40% of children. In addition, culture results take 2–6 weeks on liquid media and up to 12 weeks on solid (Lowenstein–Jensen) media to become positive, so these results are not usually available before decision-making on starting ATT [Table 1]. However, as TB affects the most vulnerable of children including the malnourished and those from poorer socioeconomic backgrounds who are more likely to be adversely affected by delayed treatment, it is important to send these investigations both for later confirmation of disease and for susceptibility testing.
The only tests other than the AFB smear that may be considered as point-of-care tests in the diagnosis of TB are nucleic acid amplification tests or TB polymerase chain reaction tests such as the Xpert MTB/Rif assay.
The Xpert MTB/RIF assay has proved to be nearly as good as the AFB culture, identifying twice as many cases as the AFB smear and 70%–90% of culture-positive TB  as well as 10%–15% of culture-negative clinically defined TB. The other advantage of the Xpert MTB/RIF assay is that a result is available within 2 h of testing and gives information as to whether the isolate is resistant to RIF which is an indicator of multidrug-resistant TB, allowing the clinician to start second-line drugs much earlier than would be possible if one were to rely only on the results of culture testing. Later, generation assays hold the promise of a bedside point-of-care test that, in addition to rapid results, will also provide information on resistance to first- and second-line antituberculous drugs. However, it must be remembered that the diagnosis of TB in children, with their attendant problems of sample collection and low numbers of bacilli giving rise to negative test results, will still rest on a clinical diagnosis in a large number of cases.
TB in children is predominantly paucibacillary – sputum smear for AFG and culture have poor diagnostic yields.
The Xpert MTB/RIF test is rapid and detects twice as many cases as AFB smear and 70%–90% of culture-positive TB.
| Starting Antituberculous Therapy in the Child With Suspected Pulmonary Tuberculosis|| |
The approach to making a decision on whether to start ATT in suspected pulmonary TB in children is given in the WHO algorithm as shown in [Figure 10].
|Figure 10: Algorithm for initiating antituberculous therapy for pulmonary tuberculosis in children (based on the WHO guidelines).|
Click here to view
In a child with persistent symptoms suggestive of TB, the decision on treating for TB rests on either a positive AFB smear or (if smear negative) the presence of two of the following: A positive contact history, physical signs suggestive of TB, or a chest X-ray suggestive of TB. If only one of the above is present, one should investigate for other causes of the child's symptoms or review later if the symptoms or signs resolve with other therapy, such as antibiotics.
Current Revised National Tuberculosis Control Program recommendations for ATT in all newly diagnosed smear-positive or smear-negative pulmonary TB are to give a 2-month intensive phase with four drugs, isoniazid, RIF, pyrazinamide, and ethambutol, followed by a 4-month continuation phase during which only pyrazinamide will be stopped [Appendix 2].
Case scenario (continued)
In our child, we suspected TB because of persistent fever and a documented weight loss of half a kilogram over 2 weeks. She has symptoms (persistent fever) and physical signs (weight loss) suggestive of TB, and a chest X-ray showing a hilar lymph node which is the hallmark of primary TB in children. These make up two of the features in the algorithm as shown in [Figure 10], and she is in contact with a close household contact who is likely to have TB when his workup is done. In this situation, even though she is smear negative, the physician is justified in starting ATT as she fulfills the two criteria for treating TB in smear-negative disease, as shown in the algorithm in [Figure 10].
When starting empiric antituberculous therapy in children, it is essential to monitor the treatment response.
- A poor clinical response to treatment may indicate poor adherence to therapy
- Incorrect doses of antituberculous drugs
- Presence of TB resistant to first-line drugs, or
- An incorrect diagnosis of TB in the first place.
When starting empiric antituberculous therapy in children, it is essential to monitor the treatment response. If the child has TB, he or she should have resolution of symptoms, such as resolution of fever and/or cough and should have started gaining weight by the end of the intensive phase (the first 2 months) of therapy (as chest X-ray findings often take several months to resolve and may leave sequelae such as fibrosis or bronchiectasis, radiographic resolution should not be used as a criterion to judge whether therapy is successful). A poor clinical response to treatment may indicate (1) poor adherence to therapy, (2) incorrect doses of antituberculous drugs, (3) presence of TB resistant to first-line drugs, or (4) an incorrect diagnosis of TB in the first place.
| Conclusion|| |
It is often difficult to confirm TB in children but not so difficult to make a clinical diagnosis of TB in a child. An accurate clinical diagnosis of TB is possible in most children with a careful history and clinical assessment as well as focused testing for TB including a chest X-ray.
Chest x-rays - courtesy Department of Radiodiagnosis, Christian Medical College, Vellore.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
TB India 2016. Revised National TB Control Programme Annual Status Report. Central TB Division, Directorate General of Health Services, Ministry of Health and Family Welfare, Nirman Bhavan, New Delhi. Available from: http://www.tbcindia.nic.in/showfile.php?lid=3180
. [Last accessed on 2016 Dec 06].
Marais BJ, Gie RP, Obihara CC, Hesseling AC, Schaaf HS, Beyers N. Well defined symptoms are of value in the diagnosis of childhood pulmonary tuberculosis. Arch Dis Child 2005;90:1162-5.
Grzybowski S, Barnett GD, Styblo K. Contacts of cases of active pulmonary tuberculosis. Bull Int Union Tuberc 1975;50:90-106.
Marais BJ, Gie RP, Schaaf HS, Starke JR, Hesseling AC, Donald PR, et al.
A proposed radiological classification of childhood intra-thoracic tuberculosis. Pediatr Radiol 2004;34:886-94.
FJW Miller. Tuberculosis in Children. Edinburgh: Churchill Livingstone; New York. 1982.
Nhu NT, Ha DT, Anh ND, Thu DD, Duong TN, Quang ND, et al.
Evaluation of Xpert MTB/RIF and MODS assay for the diagnosis of pediatric tuberculosis. BMC Infect Dis 2013;13:31.
Sekadde MP, Wobudeya E, Joloba ML, Ssengooba W, Kisembo H, Bakeera-Kitaka S, et al.
Evaluation of the Xpert MTB/RIF test for the diagnosis of childhood pulmonary tuberculosis in Uganda: A cross-sectional diagnostic study. BMC Infect Dis 2013;13:133.
Gomez-Pastrana D. Diagnosis of pulmonary tuberculosis in children. J Infect Dis Ther 2013;1:17-24.
Jain SK, Ordonez A, Kinikar A, Gupte N, Thakar M, Mave V, et al.
Pediatric tuberculosis in young children in India: a prospective study. Biomed Res Int 2013;2013:783698.
Marais S, Thwaites G, Schoeman JF, Török ME, Misra UK, Prasad K, et al.
Tuberculous meningitis: a uniform case definition for use in clinical research. Lancet Infect Dis 2010;10:803-12.
Cruz AT, Starke JR. Clinical manifestations of tuberculosis in children. Paediatr Respir Rev 2007;8:107-17.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10]