Print this page
Friday, December 23, 2016 Written by Ellen Jo Baron, Ph.D., D(ABMM), Prof. Emerita, Stanford University Director of Medical Affairs, Cepheid

The Need for Better Diagnostic Tests for Pediatric Tuberculosis

Almost unfathomable to contemplate is the fact that one-third of the world's population is infected with tuberculosis (TB). Each year approximately 9 million new infections occur (most people who become infected do not exhibit symptoms of disease for many years, if ever) but around 1.6 million people die of their disease; approximately one death every 9 seconds. 80% of the world's cases occur in 22 high burden countries, identified by the World Health Organization (WHO) (Figure 1). Tuberculosis can be cured, especially if drug-susceptible and drug-resistant strains are detected early and treated appropriately. A keystone of the WHO's global plan to stop TB ("Stop TB Partnership"; involves using smear microscopy to identify patients and to treat those patients with a course of antibiotics, with the patient being observed taking his/her drugs by a healthcare worker for at least the first two months. This "directly observed therapy short-course" (DOTS) regimen includes the oral drugs rifampin, isoniazid (INH), pyrazinamide, and ethambutol for 2 months, followed by 4 more months of INH and rifampin alone; and for patients with susceptible strains, >95% are cured. If the organism is known to be susceptible to rifampin and INH through drug susceptibility testing (DST) performed on organisms isolated in pure culture, then ethambutol may be removed from the regimen. An expanding group of patients infected with multi-drug resistant TB (MDR-TB; i.e., strains resistant to multiple drugs including both INH and rifampin) require longer therapy with more antimicrobial agents, typically fluoroquinolones, aminoglycosides, p-aminosalicylic acid, ethionamide, and others, some of which must be injected. The sooner a patient harboring a resistant strain begins taking the drugs that will actually work, the greater the likelihood that the patient will be cured, even though the therapy must continue for a much longer time than if the strain were susceptible.

Although the international DOTS strategy has been widely implemented, its overall impact on the prevalence and incidence of tuberculosis, particularly in countries with a high burden of HIV patients, has not met expectations. One reason is that even those patients on effective therapy remain infectious longer than previously realized.1 Another reason is that the side effects of anti-tuberculosis drugs are not trivial. The drugs can cause hepatitis, rashes, malaise, nausea, thrombocytopenia, and neuropathy. In addition, some of the pills are enormous and hard to swallow (Figure 2). No wonder patients must be observed taking their medications; if not being observed they would avoid taking them. The WHO Tuberculosis Fact Sheet states that although the incidence of TB was falling in all areas of the world, the population growth in some areas offsets that small victory to result in a net increase in the number of infected persons in Africa, the Eastern Mediterranean, and South-East Asia (

Recent estimates indicate that infants and children comprise 15-20% of cases of tuberculosis disease in resource-limited countries.2 In some places this may be as high as 50%.3 In developing regions, an increasing proportion of these children are co-infected with HIV, which results in more severe TB disease and higher mortality.4 Children often acquire their disease from infected adults in their own family or in the community.5 In fact, the incidence of TB in children, who usually present with symptoms within a year of infection, provides a snapshot of transmission within a community.6 Diagnosis of tuberculosis in children, however, is a major challenge. Infants and young children may cough, but they cannot be instructed to expectorate the sputum that they produce. If they cough, they usually swallow the sputum. That is the premise on which the collection of gastric aspirates is based: the swallowed acid fast bacilli (AFB) can be recovered from the stomach, especially in the morning before the stomach contents are emptied into the gastrointestinal tract. At least one study has shown that stool (ultimate destination of swallowed sputum) is another potentially useful specimen for detection of TB in children, although others have not had such positive results.7, 8

But recovering AFB from respiratory secretions regardless of the site from which they can be collected is more difficult in children. At most only 15% of children with TB have smear-positive respiratory samples.9 Their organisms tend to stay in the perihilar lymph nodes and rarely rupture into the bronchial tree, as they do in adults. Consequently, children's respiratory samples may have few or no mycobacteria to detect regardless of the system used. Children older than 8 years old are more likely to exhibit adult-like pulmonary TB with coughing and culture-positive sputum. Thus today, diagnosis of active tuberculosis in children, particularly those <8 years old, is based on a combination of clinical, radiological, and laboratory criteria. A large analysis looking at nine different algorithms for diagnosis of childhood TB, however, found major differences in the number of cases defined by each algorithm, from a low of 7% to a high of 89%!10 Criteria evaluated in various algorithms include history of contact with a TB-positive patient, HIV status, tuberculin skin-test results, chest radiographs, coughing or other respiratory symptoms, fever, weight loss, malaise, and smear and culture results from different sample types. Given the lack of a gold standard, the true endpoint cannot be known for certain. This makes evaluation of any organism-based test, such as molecular tests for genomic sequences, extremely difficult.

Cambodia is one of the world's poorest nations. Bordering Thailand and Vietnam, Cambodia is still recovering from the devastating reign of the Khmer Rouge and their ruthless leader, Pol Pot. Sixty percent of the population is below 20 years old, and it is the world's only country with a growing mortality rate in children <5 years old. It is this setting where Dr. Anne Goldfeld has chosen to try to make things better.

Dr. Anne Goldfeld, Professor in the Department of Immunology and Infectious Diseases at Harvard University, has been actively working in the field of pediatric infectious diseases for many years (Figure 3). As a principle investigator on a number of pediatric tuberculosis diagnosis and treatment projects around the resource-limited world, she works with the Cambodian Health Committee (, a subsidiary of the Global Health Committee, whose mission is to improve the health and well-being of patients, primarily children, with tuberculosis and AIDS. The work started in Cambodia. Central to these efforts is the Maddox Chivan Children's Center in Phnom Penh, Cambodia, which was built in part with grants from Angelina Jolie and Brad Pitt. Dr. Goldfeld's work was recently featured on ABC News. (See link:

AERAS (from the Greek word for "air") is another non-profit organization dedicated to ridding the world of the scourge of tuberculosis by developing an effective vaccine that is affordable and available throughout the world. Since 1921, the only vaccine for TB has been BCG (Bacillus Calmette-Guérin), named after the original investigators who developed it in Lille, France, i.e., Calmette and Guérin. In the Pasteur Institute museum in Lille one can see test tubes containing some of the actual dried-out cultures of Mycobacterium bovis on Lowenstein-Jensen agar worked on by Calmette and Guérin. Dr. Anne Goldfeld and Dr. Rinn Song, a Fellow in Infectious Diseases from Children's Hospital in Boston, are spearheading a project in the far south-eastern Cambodian province of Svay Rieng, bordering Vietnam, funded partially by AERAS. The objectives are to characterize the nature of tuberculosis disease in the local children and to determine which method of diagnosis would be best for monitoring the effectiveness of the new vaccine. As many as 1000 children will be studied in the vaccine pilot project. Several diagnostic tests are being evaluated, including an interferon-gamma releasing assay, acid fast smears, cultures in liquid and on solid media, and the Xpert® MTB/RIF (CE-IVD), which was recently endorsed by the World Health Organization. The GeneXpert Test was developed with support from the Foundation for Innovative New Diagnostics (FIND) and the National Institute for Allergy and Infectious Diseases (NIAID); the first large multi-center trial of the system was published recently in the New England Journal of Medicine.11

Children with symptoms of respiratory disease are brought to the provincial hospital in Svay Rieng (Figure 4) by their parents. They have been promised free diagnostic tests, for which they must stay overnight for 2 nights. All of their food and bedding is provided during the stay, and each parent receives a few personal hygiene products and a new sarong for the child. The children may not appreciate these sarongs, as they are used to wrap their arms and legs tightly to prevent them from squirming during the specimen-collection procedures that they must endure. At the hospital, they are first evaluated by a physician for symptoms. Radiographs are obtained and blood is drawn for laboratory tests. Urine is obtained for additional laboratory tests. A tuberculin skin test is placed, and the child and parent stay overnight in a nice room with colorful, child-size hammocks (Figure 5). The first gastric aspirate is obtained the next morning. The doctors and nurses at the district hospital are highly skilled and well-trained. The procedures they follow for specimen collection are the same or better than are performed in the best children's hospitals throughout the world.

During the gastric aspirate collection procedure, the child's oxygen saturation is monitored closely with a transcutaneous oxygen tension monitor attached to his or her toe. If the child starts breathing too shallowly and the oxygen tension drops too much, the procedure is stopped. The parent is there at the head of the bed to hold and comfort the child (Figure 6). The child is wrapped tightly in the sarong and laid on the bed. Nurses measure the length of tubing to be used and then carefully insert the nasogastric tube all the way into the stomach. Using a syringe, they withdraw the contents of the stomach and expel it into a 50 cc centrifuge tube. The markings on the tube are used to measure the volume of material recovered; if less than 5 ml, the nurse will instill sterile saline into the stomach and withdraw the liquid. It may take several syringe aspirations to reach the 5 ml required volume. The aspirate is immediately buffered to maintain a neutral pH and the specimen is placed on ice packs for transport to the laboratory in Phnom Penh, which takes place each afternoon.

Later on the second day, the children come back to the treatment rooms for collection of an aerosol-induced sputum specimen. This procedure is at least as uncomfortable as the gastric procedure. The child is first given a bolus aerosol puff of albuterol to open the lung airways (Figure 7). The parent holds the child on his or her lap for the next 15 minutes while the child breathes in the saline mist, guaranteed to irritate the lung lining and provoke coughing. The child is then wrapped in the sarong again, the oxygen monitor is placed, and the endotracheal tube is threaded down the patient's trachea about halfway to the bronchial separation. One end of the tubing is attached to a Lukens trap and the trap is attached to an electric vacuum-suction device (Figure 8). Sputum provoked by the saline mist is collected in the trap (Figure 9). Other samples for future testing, including stool specimens, are collected on this day. On the third day, the child is again brought to the treatment room early in the morning for collection of the second gastric aspirate. Will the patient and his parents ever want to see that sarong again? The skin test result is interpreted and recorded and the patient is free to go home.

What happens with those precious samples? Each specimen is labeled with at least two identifiers, and secured in a cold box (Figure 10), after which the samples are transported by car to the Pasteur Institute laboratory (Laboratoire de Biologie Médicale, Institut Pasteur du Cambodge) in Phnom Penh, about 2.5 hours away. This occurs each day; transport includes a ferry ride across the Mekong River. Stools are frozen at -80°C for processing later, and the gastric aspirates and sputum samples are handled immediately. The Pasteur Laboratory uses state-of-the-art culture and identification methods, including MGIT liquid culture, Gen-Probe nucleic acid amplification assay, and the HAIN line-probe system for species identification and drug susceptibility testing (Figure 11). A portion of each sample is refrigerated for transport to the National Center for Tuberculosis and Leprosy Control (CENAT) Laboratory, the government-run TB laboratory of Cambodia, where the other testing arm of the study is performed, which includes testing with the Xpert® MTB/RIF (CE-IVD). Cepheid loaned CHC a 4-module GeneXpert® System, sent experts in to set up the instrument and conduct training (Figure 12), and is providing all of the cartridges needed to test both gastric aspirates and sputum samples. A flowchart for specimen handling also was provided (Figure 13). The GeneXpert System fit easily into their existing space, is easy to use, and the results are beginning to accumulate (Figure 14). According to the previously cited ABC news report, the system is already making a difference.

Cepheid is proud to be contributing to this major pediatric tuberculosis diagnosis and treatment trial. Perhaps better diagnostic methods for this difficult-to-diagnose disease will come out of the study. In any case, many children are receiving high quality care, their parents are receiving health education, and improved wellbeing for many Cambodians is a direct benefit.


1. Fitzwater, S. P., et al. Prolonged infectiousness of tuberculosis patients in a directly observed therapy short-course program with standardized therapy. Clin Infect Dis. 51: 371-378.
2. Hesseling, A. C., et al. 2002. A critical review of diagnostic approaches used in the diagnosis of childhood tuberculosis. Int J Tuberc Lung Dis. 6: 1038-1045.
3. Marais, B. J. & H. S. Schaaf. Childhood tuberculosis: an emerging and previously neglected problem. Infect Dis Clin North Am. 24: 727-749.
4. Mwinga, A. 2005. Challenges and hope for the diagnosis of tuberculosis in infants and young children. Lancet. 365: 97-98.
5. Wood, R., et al. Tuberculosis transmission to young children in a South African community: modeling household and community infection risks. Clin Infect Dis. 51: 401-408.
6. Stefan, D. C., et al. Interferon-gamma release assays for the detection of Mycobacterium tuberculosis infection in children with cancer. Int J Tuberc Lung Dis. 14: 689-694.
7. Donald, P. R., et al. 1996. Stool microscopy and culture to assist the diagnosis of pulmonary tuberculosis in childhood. J Trop Pediatr. 42: 311-312.
8. Oberhelman, R. A., et al. Diagnostic approaches for paediatric tuberculosis by use of different specimen types, culture methods, and PCR: a prospective case-control study. Lancet Infect Dis.10: 612-620.
9. Marais, B. J. 2007. Childhood tuberculosis--risk assessment and diagnosis. S Afr Med J. 97: 978-982.
10. Hatherill, M., et al. Structured approaches for the screening and diagnosis of childhood tuberculosis in a high prevalence region of South Africa. Bull World Health Organ. 88: 312-320.
11. Boehme, C. C., et al. Rapid molecular detection of tuberculosis and rifampin resistance. N Engl J Med. 363: 1005-1015.