Secondary Logo

Journal Logo

Original Studies

Management of Pediatric Multidrug-Resistant Tuberculosis and Latent Tuberculosis Infections in New York City From 1995 to 2003

Feja, Kristina MD, MPH*†; McNelley, Erin MD; Tran, Cindy S. MD§; Burzynski, Joseph MD; Saiman, Lisa MD, MPH

Author Information
The Pediatric Infectious Disease Journal: October 2008 - Volume 27 - Issue 10 - p 907-912
doi: 10.1097/INF.0b013e3181783aca
  • Free


During the early 1990s, New York City (NYC) was one of the epicenters of multidrug-resistant tuberculosis (MDR-TB) in the United States. At the peak of the epidemic in 1992, 441 (12%) of the 3811 patients in NYC with TB were infected with MDR strains, which accounted for 61% of the nation’s MDR-TB cases.1 Since 1992, case rates of MDR-TB have fallen steadily in the United States. In NYC, cases of MDR-TB declined from 441 in 1992 to 21 in 2003, representing a 95% decrease.2 This impressive decline in cases is thought to have been achieved by improved public health programs; improved infection control measures in hospitals, homeless shelters, and correctional facilities; as well as earlier initiation of isolation and appropriate treatment, including directly observed therapy (DOT).3 DOT in this setting is defined as a patient being observed taking TB medication by a healthcare worker at a New York City Department of Health and Mental Hygiene (NYC DOHMH) chest center or by an outreach worker who meets the patient at a designated location.3,4

Cases of MDR-TB in children almost always represent primary disease and are considered a sentinel event (ie, evidence of ongoing transmission from an undiagnosed infectious source case with MDR-TB).5 To date, there have been relatively few reports describing MDR-TB in children. Case series of pediatric MDR-TB have generally been small (2–38 subjects) and often single-center studies.6–12 To date, only 1 study has reported latent TB infection (LTBI) in children resulting from MDR-resistant source cases.7 Thus, we describe children younger than 15 years of age with MDR-TB or MDR-LTBI treated in NYC in efforts to expand our understanding of the management and outcomes of such cases.


Study Design.

A retrospective study of pediatric cases of MDR-TB and MDR-LTBI, which were treated in NYC from 1995 to 2003, was performed. Institutional Review Board approval to perform this study was obtained from the NYC DOHMH and Columbia University.


During the study period, the Tuberculosis Control Program of the NYC DOHMH operated 10 chest centers located throughout the 5 boroughs of NYC staffed by pulmonary or infectious disease board-certified physicians. In 2003, 476 (42%) of 1141 patients diagnosed with active TB in NYC and 57% of the prevalent MDR-TB cases received care at NYC DOHMH chest centers. DOT was strongly encouraged for individuals with active TB, and in 2003, 89% of patients treated at the DOHMH chest centers received DOT.

In addition, the NYC DOHMH TB Control program monitored the treatment of all patients diagnosed with TB disease in NYC, regardless of the site of care. This ensured that TB treatment met acceptable standards and ensured accurate surveillance, epidemiology, and contact investigations. Furthermore, according to the NYC Health Code, all suspected and confirmed cases of TB and cases of LTBI in children <5 years of age must be reported to the NYC DOHMH. NYC DOHMH Surveillance and Central Registry staff entered all TB surveillance and treatment data into a computerized TB registry.

Study Subject Case Definitions.

Study subjects were identified by searching the NYC DOHMH TB registry. Because pediatric cases of TB are often culture-negative, cases were defined based on their treatment regimens. Cases were a child younger than 15 years of age with confirmed TB or LTBI who began treatment from January 1995 to December 2003 and received at least 1 month of therapy with one or more second-line TB drugs for MDR-TB or MDR-LTBI, or were treated with pyrazinamide and ethambutol for MDR-LTBI. Second-line TB drugs included quinolone agents (ciprofloxacin, levofloxacin, or ofloxacin), cycloserine, ethionamide, streptomycin, para-aminosalicylic acid, capreomycin, and/or amikacin. The susceptibility profile of the isolates from the subjects and/or their source case, when available, was also reviewed.

Data Collection.

The TB registry was abstracted for demographic characteristics and TB-related clinical data including clinical symptoms and signs, tuberculin skin test (TST) results, radiographic findings, smear and culture results, treatment, duration of therapy, adverse effects, and clinical outcome. The clinic charts of subjects were reviewed to corroborate and expand the TB registry data, when available. Two years after the study period (ie, from 2004 to 2005) the TB registry was searched to ascertain if any study subject had developed recurrent disease or progressed from LTBI to active TB.

Statistical Analysis.

Data were entered into a Microsoft Access 2000 database (Microsoft Corporation, Redmond, WA). Quantitative descriptors such as means and medians were calculated using Microsoft Excel 2000. Further statistical analysis was performed using SAS 8.02 (SAS Institute, Cary, NC).


Study Population.

Search of the TB registry identified 265 potential pediatric subjects treated with second-line drugs for suspected or confirmed TB disease and 181 pediatric patients treated for suspected or confirmed MDR-LTBI during the study period. Potential subjects did not fulfill inclusion criteria for the following reasons: (1) lack of verification by the NYC DOHMH as a case of active TB or LTBI,13,14 (2) treatment began before 1995, (3) age ≥15 years at the start of treatment, (4) <1 month of treatment with a second-line agent, or (5) duplicate records. Furthermore, the diagnosis of some subjects changed from LTBI to active TB. Thus, the final study population consisted of 20 pediatric subjects treated for MDR-TB and 51 subjects treated for MDR-LTBI. Overall, MDR-TB occurred in 3% of the 672 children diagnosed with TB disease during the study period.

Demographic Characteristics of Study Subjects.

The demographic characteristics of the study subjects were compared with those of all cases of pediatric TB in NYC during the study period (Table 1). The median age of children with MDR-TB versus MDR-LTBI was 2.7 years and 9.8 years, respectively. Eighteen (90%) subjects with MDR-TB were born in the United States, 1 subject was born in India, and 1 in Guinea. Of the 34 subjects with LTBI whose country of origin was known, 22 (65%) were born in the United States, 4 in the Dominican Republic, 3 in Vietnam, 2 in Peru, and one each in Trinidad and Tobago, Haiti, and Russia. When compared with the demographic characteristics of all pediatric TB cases in NYC,2,15–22 subjects with MDR-TB were more likely to be female (P = 0.049) and born in the United States (P = 0.047). Clinic charts were available for 33 (46%) of the 71 study subjects and confirmed the demographic and clinical data noted in the computerized TB registry.

Demographic Characteristics of Pediatric Patients With TB, MDR-TB, and Latent MDR-TB Infections in NYC, 1995–2003
Clinical Findings.

The TST, smear, and culture results of the study subjects are summarized in Table 2. The majority (75%, 15/20) of subjects with active TB had pulmonary disease. Extrapulmonary disease was noted in 4 (20%) of 20 subjects: 2 with lymph node, 1 with peritoneal, and 1 with bone/joint disease. The site of disease was unknown for 1 subject. HIV status was incomplete in the TB registry. Of the 11 subjects with MDR-TB whose clinic charts were available, HIV status was noted for 3; the mother of 1 subject was documented as HIV-negative, and 2 subjects had negative HIV serology.

Clinical Characteristics of Pediatric Patients Treated With Second-Line TB Drugs in NYC, 1995–2003
Susceptibility Patterns for MDR-TB and MDR-LTBI Isolates.

Among the 20 subjects with MDR-TB, 5 (25%) had positive acid-fast bacilli smears, 6 (30%) had positive cultures for Mycobacterium tuberculosis, and 13 (65%) had source case information in the TB registry. Susceptibility testing for the 19 strains (6 from case subjects and 13 from source cases) demonstrated resistance to a median of 5 drugs (range, 1–12 drugs). The susceptibility of the source case isolate was available for 47 (92%) of 51 patients with MDR-LTBI and demonstrated resistance to a median of 5 drugs (range, 3–7 drugs). All strains associated with the subjects in this study were resistant to isoniazid and rifampin and at least 1 other agent.

Duration and Completion of Treatment.

Fourteen (70%) of 20 cases of MDR-TB and 41 (80%) of 51 cases of MDR-LTBI were treated at NYC DOHMH TB centers. Of the 6 subjects with MDR-TB treated at non-DOH sites, 3 were treated at municipal hospital clinics and 3 by private doctors. Of the 10 subjects with MDR-LTBI, 4 were treated at municipal hospital clinics, 5 by private doctors, and the site of care was undocumented for 1 subject.

Treatment regimens were individualized according to susceptibility results. The proportion of subjects treated with specific TB drugs is shown in Table 3. Those with active TB were treated with an average of 4.4 drugs during the treatment course and the most commonly used second-line agents were cycloserine (80% of subjects), ethionamide (65% of subjects), and quinolone agents (60% of subjects). One subject’s strain was resistant to 10 drugs, and this subject was treated with linezolid, imipenem, and γ-interferon in addition to second-line TB drugs. The mean length of treatment with second-line agents for 16 cases with documentation of completion of therapy was 18.8 months (Table 4). Of the 4 subjects without complete documentation of therapy; 1 died, 2 left NYC before completing treatment, and 1 had incomplete medical records. However, these last 3 subjects received an average of 22.5 months of treatment. The majority (18/20, 90%) of subjects had documentation of DOT (mean 14.9 months).

Drugs Used to Treat MDR-TB and MDR-LTBI in Pediatric Patients in NYC, 1995–2003
Length of Treatment With Second-Line TB Drugs for Active TB and LTBI in Pediatric Patients in NYC, 1995-2003

Subjects with LTBI received an average of 3 drugs (range, 2–7 drugs), most commonly quinolone agents (69%), cycloserine (67%), ethionamide (49%), pyrazinamide (53%), and ethambutol (39%). The mean length of treatment for subjects with documentation of completion of treatment (n = 38) was 10.3 months (Table 4). Only 33% (17/51) of subjects with MDR-LTBI received DOT, and the average length of DOT was 9.1 month. A significantly greater proportion of subjects with MDR-LTBI who received care at a NYC DOHMH chest clinic completed treatment (88%, 36/41 subjects) compared with those who received care at a non-DOH site [(22%, 2/9 subjects) P < 0.001, Fisher exact test].


Adverse effects attributed to drug therapy could only be assessed by review of the clinic charts. Among subjects with available charts, 4 of 11 (36%) with MDR-TB, experienced adverse effects. These adverse effects were most commonly gastrointestinal, including nausea/vomiting (n = 1), loss of appetite and abdominal pain (n = 1), and elevated liver function tests (n = 1). Treatment was suspended, but not discontinued in these subjects and all 3 completed therapy. One subject experienced several adverse effects: hypothyroidism attributed to either para-aminosalicylic acid or ethionamide, blurred vision attributed to cycloserine, and hearing loss and muscle pain attributed to capreomycin. Both cycloserine and capreomycin were discontinued, but this subject completed therapy.

Among those with MDR-LTBI and available clinic charts, 8 (24%) of 22 experienced adverse effects attributed to drug therapy. Gastrointestinal toxicity was most common and included nausea/vomiting in 5 subjects, accompanied by abdominal pain in 3 subjects, and elevated liver function tests in 2. Two subjects experienced behavioral changes described as “depression and nervousness.” One subject developed a rash described as “itchy hives on the trunk and legs” and 1 subject experienced joint pain. Only those with elevated liver function tests discontinued treatment. Three additional subjects were reported in the TB registry as having discontinued treatment secondary to an adverse reaction, but these charts were unavailable for review.

Outcome of Treatment.

There was no evidence in the NYC DOHMH TB Registry that recurrent disease occurred among the MDR-TB subjects or that subjects with MDR-LTBI progressed to active disease during the study period or for 2 years thereafter.


To our knowledge, this study represents the largest case series of MDR-TB and MDR-LTBI in children in the United States published to date. MDR-TB in children remains relatively rare and there are few published reports of case rates of pediatric MDR-TB or MDR-LTBI. We found that our MDR-TB cases were more likely to be U.S.-born. This finding is similar to an earlier report in which MDR-TB cases treated in NYC from 1995 to 1997 were more likely to be U.S.-born, HIV-infected, health care workers, homeless, and intravenous drug users when compared with non-MDR cases.23 In contrast, a case series reported from California demonstrated that MDR-TB cases were more likely to be foreign-born and less likely to be homeless, HIV-positive, and/or intravenous drug users.24 Therefore, we believe that our findings in children reflect the local demographics of the adult source cases in NYC. We do not have an explanation for the differences in gender ratios observed among cases of pediatric TB in NYC and hypothesize that this finding is a result of the small sample size.

Our use of a treatment-based case definition rather than a microbiologically confirmed case definition may have led to an increased case rate when compared with other reports. In contrast with our rate of 3%, a U.S. survey conducted from 1993 to 2001 by the Centers for Disease Control and Prevention (CDC) found that 1.6% of M. tuberuclosis isolates from children <15 years of age were MDR.25 The rate of culture-confirmed MDR-TB in patients 19 years of age or younger in 2003 in NYC was 2.3%.2 In a single center study from NYC, MDR-TB accounted for <1% of all isolates in children ≤14 years of age from 1961 to 1980 and 10.3% of isolates from 1981 to 1992.5,26 It is also possible that MDR case rates in children are under-reported because of a lack of resources with which to obtain cultures.27

There are also few published reports of treatment of MDR-TB in children. These are shown in Table 5. Because of the small number of cases, different case definitions, and different treatment regimens, it is difficult to compare studies and develop standardized treatment guidelines. Treatment regimens are generally derived from those used in adult patients. It is possible that MDR-TB in children might be successfully treated without the parenteral antibiotics often used in adults, presumably as a result of a relatively lower organism burden in many children with TB disease,28 but this has not been adequately studied. Many authors have underscored the lack of pediatric preparations of TB drugs that may adversely impact compliance.28–30

Pediatric Series of Demographic and Clinical Characteristics of Pediatric MDR-TB, 1992–2003

Among adults with MDR-TB, crude mortality ranges between 23% and 37%31,32 and mortality rates are even higher in HIV-infected adults with MDR TB.33 In general, however, relatively fewer children with MDR-TB seem to be coinfected with HIV, although data on HIV status are often incomplete. Crude mortality rates in pediatric studies were low, ranging from 0% to 10% with the exception of a TB meningitis study in which 75% of the subjects were HIV infected.7–9,11,12 Recurrence rates of MDR-TB disease in adults range from 1.3% to 7%.31,32,34,35 Few data regarding recurrence rates are available in children. The rates of relapse after treatment of MDR-TB disease in pediatric series have ranged between 0% and 3%.7–9,11

Although our sample size was small, this series demonstrated that the treatment of MDR-TB and MDR-LTBI in pediatrics seemed to be effective, no cases of TB disease were reported in any of our study subjects 2–10 years after completion of treatment. To date, there are few published reports of long-term outcome data for children. Schaaf et al8 followed 21 children who had completed treatment for MDR-TB for a median of 14.8 months during which time one HIV-infected child had another episode of TB, with a drug-susceptible strain. In another study, Schaaf et al also followed 29 children with MDR-TB treated with a variety of regimens including no treatment (n = 4). At 30 months, all were clinically and “radiographically well.”7 The rate of apparent cure in our study is comparable to the 75% favorable long-term outcome reported in adults with MDR-TB.34

There are even fewer studies of MDR-LTBI in children than studies of MDR-TB due to the inherent difficulty of diagnosis, lack of consensus regarding treatment, and the challenges of documenting management. Our small series did show the potential benefit of treatment for MDR-LTBI. However, treatment was suboptimal as DOT was prescribed for only 33% of subjects and among those cared for at a non-DOH site, only 22% completed therapy. Efficacy was comparable to a previous report; in 30 months of follow-up, Schaaf et al7 reported that 2 (5%) of 41 pediatric contacts of adult MDR-TB cases who received appropriate chemoprophylaxis developed probable TB disease.

Second-line TB drugs are generally better tolerated by children than adults, although adverse reactions occur more frequently than noted with first-line drugs.30 The most common toxicities reported in children treated for MDR-TB were gastrointestinal complaints such as nausea, vomiting, and/or abdominal pain which occurred in 32–42% of patients;7,8,11 psychiatric effects that occurred in 6–11%,9,11 hypothyroidism in 6–8% of patients,9,11 and possible arthralgia that occurred in 7% of patients taking ofloxacin.8 Data regarding long-term adverse effects are lacking. Our series also demonstrated that treatment of MDR-TB and MDR-LTBI was generally well tolerated by children, but our data are incomplete as a result of missing clinic charts.

In adults with MDR-TB, quinolones have assumed a major role in treatment because of in vitro activity against M. tuberculosis, a relatively low rate of toxicity, and improved microbiologic and clinical outcomes.34 Many clinicians have been concerned about the use of quinolone agents in children because of arthropathy observed in immature puppies after long-term administration.36 However, during the past decade, there has been increasing comfort using quinolone agents in children, but these treatment courses have generally been short (eg, 15–30 days of treatment for a pulmonary exacerbation in patients with cystic fibrosis).37 Observational studies in cystic fibrosis and other pediatric patients have demonstrated a low rate (1.5–3.8%) of reversible arthralgias or myalgias similar to the rate observed in adults.37,38 Three percent (1/33) of children in our series had joint complaints while receiving quinolone therapy, but not all clinic charts were available for review.

Our study did have several limitations. There were missing data in the TB registry and unavailable clinic charts, particularly for those subjects receiving care at non-DOHMH clinics. The clinic charts were particularly important for obtaining specific drug toxicities and the reasons why therapy was discontinued. This series cannot elucidate the optimal treatment regimens for pediatric MDR-TB or MDR-LTBI as multiple providers were responsible for designing treatment regimens. We were only able to assess recurrent disease or progression from LTBI to active TB among children who remained in NYC. Finally, source case data was missing for 4 cases of LTBI and 1 case of TB. Thus, these cases may have been misclassified as MDR.

In summary, the treatment of MDR-TB and MDR-LTBI in children seems to be efficacious and generally well tolerated. Children requiring these complex treatment regimens are best cared for in treatment settings with expertise in MDR-TB that have public health supervision to ensure completion of treatment. A multicenter registry for MDR-TB and MDR-LTBI in children is desirable to obtain accurate rates of toxicity and cure.


1. Bloch AB, Cauthen GM, Onorato IM, et al. Nationwide survey of drug-resistant tuberculosis in the United States. JAMA. 1994;271:665–671.
2. Tuberculosis in New York City, 2003: Information Summary. New York, NY: New York City Department of Health and Mental Hygiene; 2004.
3. Freiden TR, Fujiwara PI, Washko RM, Hamburg MA. Tuberculosis in New York City—turning the tide. N Engl J Med. 1995;333:229–233.
4. Bureau of Tuberculosis Control New York City Department of Health. Clinical Policies and Protocols. 3rd ed. New York, NY: New York City Department of Health; 1999.
5. Steiner P, Rao M. Drug-resistant tuberculosis in children. Semin Pediatr Infect Dis. 1993;4:275–282.
6. Schluger NW, Lawrence RM, McGuiness G, Park M, Rom WN. Multidrug-resistant tuberculosis in children: two cases and a review of the literature. Pediatr Pulmonol. 1996;21:138–142.
7. Schaaf HS, Gie RP, Kennedy M, et al. Evaluation of young children in contact with adult multidrug-resistant pulmonary tuberculosis: a 30-month follow-up. Pediatrics. 2002;109:765–771.
8. Schaaf HS, Shean K, Donald PR. Culture confirmed multidrug resistant tuberculosis: diagnostic delay, clinical features, and outcome. Arch Dis Child. 2003;88:1106–1111.
9. Mukherjee JS, Joseph JK, Rich ML, et al. Clinical and programmatic considerations in the treatment of MDR-TB in children: a series of 16 patients from Lima, Peru. Int J Tuberc Lung Dis. 2003;7:637–644.
10. Soeters M, de Vries AM, Kimpen JL, et al. Clinical features and outcome in children admitted to a TB hospital in the Western Cape–the influence of HIV infection and drug resistance. S Afr Med J. 2005;95:602–606.
11. Drobac PC, Mukherjee JS, Joseph JK, et al. Community-based therapy for children with multidrug-resistant tuberculosis. Pediatrics. 2006;117:2022–2029.
12. Padayatchi N, Bamber S, Dawood H, Bobat R. Multidrug-resistant tuberculous meningitis in children in Durbin, South Africa. Pediatr Infect Dis J. 2006;25:147–150.
13. San Gabriel P, Saiman L, Kaye K, Silin M, Onorato I, Schulte J. Completeness of pediatric TB reporting in New York City. Public Health Rep. 2003;118:144–153.
14. TB Annual Summary: 2006. New York, NY: New York City Department of Health and Mental Hygiene; 2008.
15. Tuberculosis in New York City, 1995: Information Summary. New York, NY: New York City Department of Health and Mental Hygiene; 1996.
16. Tuberculosis in New York City, 1996: Information Summary. New York, NY: New York City Department of Health and Mental Hygiene; 1997.
17. Tuberculosis in New York City, 1997: Information Summary. New York, NY: New York City Department of Health and Mental Hygiene; 1998.
18. Tuberculosis in New York City, 1998: Information Summary. New York, NY: New York City Department of Health and Mental Hygiene; 1999.
19. Tuberculosis in New York City, 1999: Information Summary. New York, NY: New York City Department of Health and Mental Hygiene; 2000.
20. Tuberculosis in New York City, 2000: Information Summary. New York, NY: New York City Department of Health and Mental Hygiene; 2001.
21. Tuberculosis in New York City, 2001: Information Summary. New York, NY: New York City Department of Health and Mental Hygiene; 2002.
22. Tuberculosis in New York City, 2002: Information Summary. New York, NY: New York City Department of Health and Mental Hygiene; 2003.
23. Munsiff S, Bassoff T, Nivin B, et al. Molecular epidemiology of multidrug-resistant tuberculosis, New York City, 1995–1997. Emerg Infect Dis. 2002;8:1230–1238.
24. Granich RM, Oh P, Lewis B, Porco TC, Flood J. Multidrug resistance among persons with tuberculosis in California, 1994–2003. JAMA. 2005;293:2732–2739.
25. Nelson LJ, Schneider E, Wells CD, Moore M. Epidemiology of childhood tuberculosis in the United States from 1993–2001: the need for continued vigilance. Pediatrics. 2004;114:333–341.
26. Steiner P, Rao M, Victoria MS, Hunt J, Steiner M. A continuing study of primary drug-resistant tuberculosis among children observed at the Kings County Hospital Medical Center between the years 1961–1980. Am Rev Respir Dis. 1983;128:425–428.
27. World Health Organization. Antituberculosis Drug Resistance in the World, Report No. 3. 2004.
28. Starke JR, Jacobs RF, Jereb J. Resurgence of tuberculosis in children. J Pediatr. 1992;120:839–855.
29. Smith KC, Seaworth BJ. Drug-resistant tuberculosis: controversies and challenges in pediatrics. Expert Rev Anti Infect Ther. 2005;3:995–1010.
30. Swanson DS, Starke JR. Drug-resistant tuberculosis in pediatrics. Pediatr Clin North Am. 1995;42:553–581.
31. Goble M, Iseman MD, Madsen LA, Waite D, Ackerson L, Horsburgh CR Jr. Treatment of 171 patients with pulmonary tuberculosis resistant to isoniazid and rifampin. N Engl J Med. 1993;328:527–532.
32. Mitnick CD, Bayona J, Palacios E, et al. Community-based therapy for multidrug-resistant tuberculosis in Lima, Peru. N Engl J Med. 2003;348:119–128.
33. Fischl MA, Daikos GL, Uttamchandani RB, et al. Clinical presentation and outcome of patients with pulmonary tuberculosis caused by multiple-drug resistant bacilli. Ann Intern Med. 1992;117:184–190.
34. Chan ED, Laurel V, Strand MJ, et al. Treatment and outcome analysis of 205 patients with multidrug-resistant tuberculosis. Am J Respir Crit Care Med. 2004;169:1103–1109.
35. Tahaoglu K, Torun T, Sevim T, et al. The treatment of multidrug-resistant tuberculosis in Turkey. N Engl J Med. 2001;345:170–174.
36. Grady R. Safety profile of quinolone antibiotics in the pediatric population. Pediatr Infect Dis J. 2003;22:1128–1132.
37. Hampel B, Hullmann R, Schmidt H. Ciprofloxacin in pediatrics: worldwide clinical experience based on compassionate use–safety report. Pediatr Infect Dis J. 1997;16:127–129; discussion 160–162.
38. Chalumeau M, Tonnelier S, D’Athis P. Fluoroquinolone safety in pediatric patients: a prospective, multicenter, comparative cohort study in France. Pediatrics. 2003;111(6 Pt 1):e714–e719.

multidrug-resistance; tuberculosis; latent TB infection; pediatrics; New York City

© 2008 Lippincott Williams & Wilkins, Inc.