Countering "Urban Myths" About Pediatric Tuberculosis

I am writing in reference to “Diagnostic Challenges Associated With Pediatric Tuberculosis: 2 Cases,” (Consultant For Pediatricians, August 2009, page 300).  At our pediatric TB Clinic, we try to counteract the various “urban myths” about TB that have been inculcated into the residents on their other rotations. We also provide our residents with hopefully a better understanding of the pathophysiology of TB in children so that they can make more educated risk/benefit decisions about the workup and treatment of patients.

I believe that many of the practicing pediatricians and other primary care practitioners have a level of knowledge that is similar to that of our arriving residents because most of them were exposed to the same information. We find that a great deal of our residents’ initial confusion is the result of terminological imprecision that leads to fuzzy thought processes.

An article such as this is hopefully written in an effort to convey the basic concepts in a simple, logical and accurate manner-and where a particular facet of a case deviates from what might be considered “best practice,” to explain the nature of the difference so that the reader receives a clear take-home message. I believe that the first case presented falls short in several respects. I would not want other physicians to get the impression that this patient’s diagnostic workup represents a standard of practice. I have therefore provided the following critique.

--Daniel Essin, MD, FAAP, FCCP
Attending Physician
Pediatric Tuberculosis Clinic, Los Angeles County
University of Southern California Medical Center

We appreciate Dr Essin's willingness to share his expertise.  The case report was written from a General Pediatrician's perspective with the purpose to state the actual events and management of an interesting case, and then make teaching points of presentation and management of TB in infancy.  Our responses to Dr Essin's comments follow.

--Sharda Udassi, MD
Assistant Professor
Division of Critical Care
University of Florida College of Medicine, Gainesville

Dr Essin comment:  In the second paragraph of Case 1 it states, “There was no history of fever with chills, night sweats, nausea or vomiting.” Chills are defined as "a sensation of cold occurring in most fevers." A rigor is "a profound chill with pilo-erection associated with teeth chattering and severe shivering."¹ These are often the symptoms reported by adults experiencing reactivation pulmonary TB; however, they are not the symptoms of primary TB in children, and therefore are not pertinent negatives. REFERENCE:
1. Isselbacher KJ, ed. Harrison's Principles of Internal Medicine. 13th ed. New York: McGraw-Hill Inc; 1994: 83.

Dr Udassi response:  Good teaching point.  Typically the very young may have “constitutional symptoms, such as fever, fussiness, weight loss.”  Hemoptysis is from caviativary lesions that the very young typically do not have.  Sweats and chills may be seen in older children.¹ 

REFERENCE:
1. Lighter J, Rigaud M. Diagnosing childhood tuberculosis: traditional and innovative modalities. Curr Probl Pediatr Adolesc Health Care. 2009;39:61-88.

Dr Essin comment:  The finding of adenopathy, airway compression, nonuniform hyperinflation and constitutional symptoms in a 10-week-old is almost pathognomonic of primary TB. It is not clear from the case whether a lateral chest radiograph was obtained. It is possible that hilar adenopathy would have been visible in this view, and a CT scan could have been avoided.


Dr Udassi response:  Ideally, the diagnosis would have been made with clinical symptoms and the chest radiograph. However, when the diagnosis of TB is in question, CT has been recommended by many groups.^1-3

In addition to TB, the differential diagnoses for hilar adenopathy in infants includes sacoidosis, lymphoma, viral infections, Mycoplasma pneumonae infection, and fungal infections (histoplasmosis, blastomycosis).

REFERENCES:
1. Wong KS, Huang YC, Lin TY. Radiographic presentation of pulmonary tuberculosis in young children. Acta Paediatr Taiwan. 1999;40:171-175.
2. Andronikou S, Joseph E, Lucas S, et al. CT scanning for the detection of tuberculous mediastinal and hilar lymphadenopathy in children. Pediatr Radiol. 2004;34:232-236.
3. Neu N, Saiman L, San Gabriel P, et al. Diagnosis of pediatric tuberculosis in the modern era. Pediatr Infect Dis J. 1999;18:122-126.

Dr Essin comment:Bronchoscopy is contraindicated and potentially lethal in a case like this, especially at this early stage of the evaluation and especially in an infant that is in respiratory distress. The risks are high and include laryngospasm, bronchospasm and desaturation and the provocation of additional airway edema that would further compromise the patient’s respiratory status.^1,2 The information that was obtained from the bronchoalveolar lavage (BAL) fluid could have been obtained with an equal probability of success by performing repeated gastric lavage since any organisms present in the sputum will eventually arrive in the stomach.REFERENCES:
1. de Blic J,  Marchac V,  Scheinmann P. Complications of flexible bronchoscopy in children: prospective study of 1,328 procedures. Eur Respir J .2002;20:1271-1276.
2. Hannania S, Barak M, Katz Y. Unilateral negative-pressure pulmonary edema in an infant during bronchoscopy. Pediatrics. 2004;113;e501-e503.

Dr Udassi response:   Some case reports suggest bronchoscopy could be helpful both diagnostically and clinically.^1-3  Clinical improvement after airway debridement can occur and was noted in our patient. 

REFERENCES:
1. Cakir E, Uyan ZS, Oktem S, et al. Flexible bronchoscopy for diagnosis and follow up of childhood endobronchial tuberculosis.  Pediatr Infect Dis J. 2008;27:783-787.
2. Donato L, Helms P, Barats A, Lebris V. Bronchoscopy in childhood pulmonary tuberculosis
[Article in French]. Arch Pediatr. 2005;12(Suppl 2):S127-S131.
3. Ledesma Albarrn JM, Prez Ruiz E, Fernndez V, et al. Endoscopic evaluation of endobronchial tuberculosis in children. [Article in Spanish] Arch Bronconeumol. 1996;32:183-186.
 

Dr Essin comment:  I disagree with the authors that a tuberculin test showing 10-mm induration "could be" interpreted as positive in this case.  I think it should be interpreted as positive.   According to American Academy of Pediatrics (AAP) Red Book, a 10-mm tuberculin skin test result should be considered positive in the following setttings:

•When children have been in close contact with persons suspected of being contagious (such as the recently incarcerated).
•When children are from a region where the prevalence of TB is high, 10 mm is considered positive even in the absence of other risk factors.

The region of Florida from which this case report originated is in or near a high prevalence area.* In addition, the father had recently been incarcerated.  The case report does not state whether the infant had contact with any of the father's associates (perhaps also recently incarcerated).


*In 2007, Florida had the 4th highest number of TB cases of all 50 states, with only California, New York, and Texas having more cases of TB. From 2005 to 2006, the number of TB cases increased in Miami and Jacksonville. Florida has the 3rd highest number of TB cases among residents of correctional facilities and among individuals who are homeless, following California and Texas. http://www.cdc.gov/tb/statistics/reports/2007/pdf/fullreport.pdf and other internet sites.

Dr Udassi response:  This is a reasonable point, although one could argue that recent incarceration does not equal “suspected of being contagious.”  After the submission of the case report, we discovered the infant had been exposed to an adolescent with TB who was at the WIC clinic the same time the infant was.  This made transmission from the patient's father or father’s associates rather unlikely. 

Dr Essin comment:  The authors write that “The diagnosis of infant TB is complicated by the lack of a clear response to a tuberculin skin test (PPD), the absence of a gold standard diagnostic test, and the difficulty in collecting respiratory specimens." In infants who present with hilar and/or mediastinal adenopathy and evidence of segmental or nonuniform airway compromise, such as air trapping and respiratory distress, primary tuberculosis is almost always the cause. With a sufficiently high index of suspicion, as in a case like this, a zero mm PPD would not alter the diagnosis. As to the difficulty of collecting specimens – collecting gastric aspirates is not difficult, although they do not yield any organisms in one-half to three-quarters of the cases. If a culture result is available from the adult source case, pursuit of these specimens is unnecessary.Dr Udassi response:  I agree with Dr Essin's comments; however, in this infant, no confirmed adult source case was found at that time – months later, we were able to find the adolescent exposure at the WIC office.

Dr Essin comment:  Gastric aspirates do not have to be collected over 3 consecutive days in the fasting state. In infants and young children, we have had good results by collecting an early morning fasting specimen leaving the nasogastric tube in place, give some clear liquids, wait several hours, collect another specimen and repeat the procedure for a third specimen. We do this all in an outpatient setting; no hospitalization is required. The early AM regimen is a throwback to the days when children with TB or suspected TB were hospitalized for extended periods and the every morning schedule was a natural part of the hospital routine.Dr Udassi response:   Interesting findings.  Typically, early morning aspirates are still recommended, as was noted in a review in Seminars in Pediatric Infectious Disease.  Note that a recent South African study suggested doing multiple sites, same day chest radiographs; however, the South African study could be complicated by high prevalence of HIV.

See also Lobato MN, Loeffler AM, Furst K, et al. Detection of Mycobacterium tuberculosis in gastric aspirates collected from children: hospitalization is not necessary. Pediatrics. 1998;102:e40

Dr Essin comment:  Regarding treatment, in infants with this degree of airway compromise, it is not uncommon for the lymph node swelling to increase as the anti-TB treatment is begun. Therefore it is advisable to either pretreat with steroids (for 1-2 days), administer steroids concomitant with the anti-TB therapy or at least observe carefully for signs of worsening distress – depending on the degree of initial compromise. This case report does not indicate whether steroids were continued after the TB drugs were started or whether the degree of respiratory distress was affected when they were stopped.Dr Udassi response:  Multiple studies suggest possible usefulness of steroids in this setting.  Some experts would not want to give steroids without concurrent TB medications.  Other articles, do not specify the use of steroids as routine for all TB.¹REFERENCE:
1. Inge LD, Wilson JW. Update on the treatment of tuberculosis. Am Fam Physician. 2008;78:457-465, 469-470.

Dr Essin comment:  I disagree with the statement “early and proper treatment minimizes the risk of multidrug-resistant TB…”  Early treatment, per se, has nothing to do with preventing the development of emergence of resistant organisms. What is important is proper treatment (ie, at least 2 drugs to which the organism is known or believed to be sensitive). Because this information is rarely available at the time treatment is begun, this will require that treatment be initiated with at least 3 drugs, the choice of which must be guided by the patterns of resistance prevalent in the community.

In this case, the problem that required early treatment was the airway compromise and respiratory distress. Prolonged treatment with steroids may be required to limit airway compromise, because it may require weeks or months after the TB treatment has begun before the swelling of the lymph nodes subsides.Dr Udassi response: Agree, it is the proper treatment only.

Dr Essin comment:  Regarding the statement "The patient was not placed in a negative-pressure room…”  As mentioned above, TB is rarely transmitted by individuals who do not have a large population of organisms in direct communication with the airways. This is the reason that the infant did not require isolation in the hospital, not that infants have a low FEV₁ and tidal volume. There is simply an insufficient concentration of organisms present in the bronchial secretions at any one time to render the patient communicable. Patients are not considered to be significantly communicable if their sputum smears are negative. It takes at least 1000 - 50,000 organisms per cc of sputum for acid-fast organisms to be discovered when the smear is examined (depending on the technique of preparation and skill of the technician). The fact that a few organisms can be recovered from BAL fluid or gastric washings is not sufficient evidence that the patient poses a risk to others.

In a patient that is smear-positive and/or known to be communicable, a negative-pressure room alone is not a sufficient isolation measure. Susceptible individuals who enter the patient’s room must wear effective personal protective devices.Dr Udassi response: Agree.

Dr Essin comment:  Regarding liver function... The drugs used to treat TB have a very low incidence of hepatotoxic adverse effects in infants and young children. The AAP Red Book does not recommend routine monitoring of liver enzymes in otherwise healthy children. Such monitoring is advised for the first several months of treatment in children with severe TB. In any case, the reason for monitoring is not that the drugs are potentially toxic to adults, it is that they are potentially toxic to anyone.
Dr Udassi response: Agree, monitoring is not done routinely for all patients, but TB drugs can be potentially toxic to anyone and monitoring is advised and was done for this patient, in first few months, as mentioned above.

Dr Essin comment:  Baseline and follow-up ophthalmological evaluations is supported by neither experience nor by the literature. A recent summary by the Royal College of Ophthalmologists¹ concluded that “Given the capricious nature of vision failure from ethambutol and its extreme rarity in children, a screening method for children seems unnecessary and impractical.” This is especially true in infants who cannot cooperate with the required examination, which, as the authors point out, requires the patient to report visual defects and blurred vision during the examination. The same point is made by Graham and colleagues.²

Two of their key messages are:
• Ethambutol toxicity is generally dose-related (the risk of optical neuritis at 15 mg/kg/d is now considered to be negligible).
•There is no confirmed report of ethambutol toxicity in children, except perhaps in TB meningitis. REFERENCES:
1. Jones DH, Russell-Eggitt I. Is it necessary to screen children for ethambutol toxicity? Recommendations for Clinical Surveillance. http://www.rcophth.ac.uk/docs/publications/paed-patient-information/EthambutolJuly2008.pdf2. Graham SM, Daley HM, Banerjee A, et al. Ethambutol in tuberculosis: time to reconsider? Arch Dis Child. 1998;79:274-278.

Dr Udassi response: Generally agree. Some experts still would examine children 5 years and older, keeping in mind that even with screening, vision failure from ethambutol use can occur quickly.^1-3 As per the statement from the Royal College of Ophthalmologists, screening in this case was advised only for the same reason that Dr Essin explained nicely above.

REFERENCES:
1. Thee S, Detjen A, Quarcoo D, et al. Ethambutol in paediatric tuberculosis: aspects of ethambutol serum concentration, efficacy and toxicity in children. Int J Tuberc Lung Dis. 2007;11:965-971.
2. Donald PR, Maher D, Maritz JS, Qazi S. Ethambutol dosage for the treatment of children: literature review and recommendations. Int J Tuberc Lung Dis. 2006;10:1318-1330.
3. Trbucq A. Should ethambutol be recommended for routine treatment of tuberculosis in children? A review of the literature. Int J Tuberc Lung Dis. 1997;1:12-15.