- Email: [email protected]
Outbreak of drug-resistant tuberculosis with second-generation transmission in a high school in California
of drug-resistant tuberculosis with neration transmission in a high school in Ren& Ridzor,,, AID, Joseph H. Kent, ~ Sarah Valway, DMD,AIPH~Penny Weismullea Dr PH, Roberta A'Ia.~ewell,PhD,.,14aryellenEleoc/c, ~'IPH,Jo~y A/ieado~;AID, Sarah Royce, MD, MP~ Ab~qail Shefer, ~¢D, Philip Smith, PhD, Charles Wood@, PhD, and Ida Onorato, AID Background: In spring 1993, four students in a high school were diagnosed with tuberculosis resistant to isoniazid, streptomycin, and ethionamide.
Methods: To investigate potential transmission of drug-resistant tuberculosis, a retrospective cohort study with case investigation and screening by tuberculin skin tests and symptom checks was conducted in a high school of approximately 1400 students. Current and graduated high-school students were included in the investigation. D N A fingerprinting of available isolates was performed.
Results: Eighteen students with active tuberculosis were identified, Through epidemiologic and laboratory investigation, 13 cases were linked; 8 entered 12th grade in fall 1993; 9 of 13 had positive cultures for Alycobacterium tuberculosis with isoniazld, streptomycin, and ethionamide resistance, and all 8 available isolates had identical D N A fm:
In the United States drug-resistant tuberculosis has become a public health problem.1 Most reported outbreaks of drug-resistant TB have occurred in health care settings and have involved primalqly human immunodeficiency virus-infected patients. 2-5 Because of congregate conditions and extensive contact among students, schools are environments where transmission of TB occurs. 6"14 In this report we describe findings of a large outbreak investigation of drug-resistant TB in a California high school with approximately 1400 students.
gerprints. No staff member had tuberculosis. One student remained infectious for 29 months, from J a n u a r y 1991 to June 1993, and was the source case for the outbreak. Another student was infectious for 5 months before diagnosis in May 1993 and was a treatment failure in February 1994 with development of rifampin and ethambutol resistance in addition to isoniazid, streptomycin, and ethionamide. In the fall 1993 screening, 292 of 1263 (23%) students tested had a positive tuberculin skin test. Risk of infection was highest among 12th graders and classroom contacts of the two students with prolonged infectiousness. An additional 94 of 928 (10%) students tested in spring 1994
METHODS Epidemiologic Investigation
had a positive tuberculin skin test; 22 were classroom contacts of the student with
DEFINITIONS
treatment failure and 21 of these had documented tuberculin skin test conversions.
Case finding was conducted from the period of start of symptoms of the first outbreak case until the end of our investigation. A case of TB was defined as a current or former high-school student with signs and symptoms compatible with TB, diagnosis between January 1, 1991, and June 30, 1994, and laboratory evidence of TB or a positive tuberculin skin test. Laboratory evidence of TB included a clinical specimen that was smear positive for acid-fast bacilli, histologic evidence of TB, or culture positive for AlycobacterizLm ttaSerculasis. A case of multidrug-resistant TB was defined as a TB case with an isolate resistant to at least isoniazid and rifampin.
Conclusion: E~*ensive transmission of drug-resistant tuberculosis was documented in this high school, along with missed opportunities for prevention and control of this outbreak. Prompt identification of tuberculosis cases and timely interventions should help reduce this public health problem. ( J Pediatr 1997; 131:863-8)
From the Epidemic lntell~qeneeService, Epidemiology Program Office, Division of Tuberculosis E/imbmtion, Center, for Disease Controland Prevention, Atlanta, Georgia;Orange Count. Hea[rb CareAgency, Santa A~vz, California; California Department of Health Services, Berkeley;Dids hm of AIDS, STD, and TB Laboratmy Research, Centersfor Di~e,we Control and Prevention,Atlanta.
Submitted for publicationAug. 6, 1996; accepted Mar. 20, 1997. Reprint requests: Ren& Ridzon, MD, Divisionof TuberculosisElimination,Centers for Disease Control and Prevention, 1600 Clif}onRd., Atlanta,GA 30333. Copyright© 1997by Mosby-YearBook, Inc. 0022-3476/97/$5.00 ~ 0 9/21/82112
863
THE JOURNAL OF PEDIATRICS DECEMBER 1997
RIDZON ET AL.
Cases 6
•
Outbreak-relatedc a s e
[ ~ Nonoutbreak-related c a s e
Symptom Onset for Source case
1
Source Case Diagnosed with TB
N Jan- Apr- dulMar Jun Sep
OctDec
Jan- AprMar Jun
1991
Jul-
Oct-
Sep
Dec
1992
Jan- AprMar Jun
Jul-
Oct-
Sep
Dec
1993
Date of Diagnosis
F/g. L Casesof TB by date of diagnosis,California High School,
For persons from whom an isolate of 3/L
tuberculosis was obtained, an outbreak-related TB case was a case whose isolate demonstrated I S E resistance and had identical D N A fingerprints. If no isolate was obtained, the case was considered outbreak-related if an epidemiologic link, defined as a household contact, friend, or classroom contact of an outbreak-related case, was found. Cases were considered potentially infectious if evidence of pulmonary disease by chest radiograph was found or if a sputum or bronchoscopy specimen was either A F B smear-positive or culture-positive for M. tuberculosis. For this investigation, the beginning of an individual's infectious period was the date of cough onset or 4 weeks before the collection date of the first culture-positive specimen, whichever was earlier. It extended until the date when the first of three consecutive specimens, obtained on separate days, was smear and culture negative. If sputa were not obtained after start of therapy, the infectious period extended until 2 weeks after appropriate antituberculous medications were started. CASE FINDING AND SCHOOL-WIDE T S T SCREENING
To find cases of active TB and TB infection at the high school, we conducted a 864
school-wide TST screening in fall 1993. Students and staff with a documented prior positive TST were not rescreened. We conducted a second school-wide screening in spring 1994 because three students attending the school were diagnosed with TB as a result of the fall screening and one student first diagnosed in M a y 1993 failed treatment and had A F B smear-positive sputum in February 1994. Each person screened was interviewed for symptoms consistent with TB and completed a self-administered questionnaire. TST screening was performed by the M a n t o u x method using 5 TU Tubersol (brand names used for identification purposes only) purified protein derivative. TSTs were applied and read 48 to 72 hours later by county health department nurses. For the purpose of providing preventive therapy for contacts of infectious cases, a positive TST was defined as >5 mm induration. Because of the large proportion (50%) of students horn in countries of high TB incidence (77% and 9% of foreign-born students were from Vietnam and Mexico, respectively) and potential positive reactions caused by bacille Calmette-Gu4rin vaccination, to increase specificity, we defined a TST of >10 mm as positive in the epidemiologic study. A TST conversion was defined as
an increase in induration of at least 10 mm in the previous 2 years. 15 All students with positive TSTs had chest radiographs perforraed. To identify classroom contacts of infectious cases, class schedules were obtained. A cohort of 32 foreign-born students was examined for the presence of a Bacille Calmette-Gu4rin scar. Additional cases of TB were Sought by matching the state and county TB registries from January 1990 through June 1994 to the school roster that contained names of staff, current students, "transfer students, and students who graduated in 1992 and 1993. County TB clinic records were reviewed for persons 13 to 20 years old, and county public health laboratoW records were reviewed to identify persons whose isolate had a susceptibility pattern that included I S E resistance. All available medical records for persons identified by these methods were reviewed.
Laboratory Investigation Susceptibility testing of ~L. tuberculosis isolates was conducted using the proportional method.16 Available isolates were sent to the Centers for Disease Control and Prevention for D N A fingerprinting using IS6110 restriction fragment length polymorphism and mixed-linker polymerase chain reaction methods. 17'18
Statistical Analysis Data were analyzed using Epi Info, version 5.0.19 For analysis of the fall 1993 TST screening data, the 9th grade (new high-school entrants not likely to have been exposed to outbreak-related cases) was used as a referent group to estimate grade-specific relative risks for a positive TST. Students in grades 10 through 12 who were new transfers into the school and therefore not exposed during the prior year were excluded from analysis of f~l screening results. Analyses were also stratified by birthplace to eliminate confounding from infections acquired outside the United States.
RESULTS During spring of 1993, two cases of active pulmonary TB were identified among 1 lth grade students. Through contact in-
~IDZON ET AL.
THE JOURNAL OF PEDIATRICS
Volume 131, Number
6
vestigations conducted by the local health department, three additional TB cases were identified in 1 lth grade students. ,4£. tuberculosi~ isolates from four of the five cases had identical susceptibility patterns with resistance to ISE. An outbreak investigation was initiated in fall 1993. ACTIVE TUBERCULOSIS IN HIGHSCHOOL STUDENTS
Cough Onset
1 I ~v~
A total of 18 TB cases were identified among current high-school students and 1992 and 1995 graduates. Cases were diagnosed from January 1991 to November 1995(Fig. 1). There were no cases among staff. Five cases were not related to the outbreak. Four were culture positive for z//L. tuberculos/a, all had drug susceptibility patterns that did not include ISE resistance, and each had a distinct D N A fingerprint. The fifth case was diagnosed on clinical criteria and did not have an epidemiologic link to any outbreak-related case. On the basis of epidemiologic and laboratory data, 13 cases were outbreak-related. Ten cases were female, 10 were Asian, and 9 were foreign born. Four cases graduated in spring 1993, eight were in 12th grade, and one was in 10th grade. Nine of these 15 cases were culture-positive; eight had d/Z. tubereu[osia isolates that were initially resistant to ISE, and one was initially resistant to ISE and ethambutol. All eight isolates available for D N A fingerprinting (including the ISE/ethambutolresistant isolate) had identical patterns. In addition to the nine culture-positive cases, one other case demonstrated AFB on biopsy specimen examination of lymphoid tissue, one had granuloma on pleural biopsy specimen, and the remaining two were diagnosed clinically. Nine had pulmonary disease, three had pleural disease, and one had lymphadenitis. Seven cases attended school[ while they had pulmonary disease. Of these seven, three had multiple AFB-srnear positive sputa, t h r e e had multiple sputum smears negative for AFB, one had a single sputum specimen with rare AFB and multiple other AFB smear-negative specimens. On the basis of medical record review and clinical course and analysis of screening data, 2 of these 1S cases were determined to be responsible for transmission
Lab report to HD
TB diagnosis
J
t 1
.......... - ........
County Sputum contacts County culture physician assumes negative
J cae J i
T Tr lllll
I/R/Z started
EMB added
Positive sputum cultures
- Ethmb~toI
Jan
Apr
Jul
Oct
Jan
Apr
I
I
I
I
I
I
1991
Jui
1992
Oct
Jan 1993
Apr
Jul
Oct
Jan 1994
F ~ . 2. Sourcecasehistory Solid bar representsthe case'sinfectiousperiod,HD, Health Department; I/R/Z, isoniazidrifampin,and pyrazinamide;EJNB,ethambutol,
during this outbreak. Their clinical courses are summarized below. SOURCE CASE
The source case entered 12th grade in fall 1993. The student was born in Southeast Asia, moved to the United States as a child, and by history had a positive TST and was treated with a 6-month course of isoniazid. Cough developed in J a n u a r y 1991 and persisted through August 1993 (Fig. 2). From Oanua W 1991 until February 1992 (while in 9th and 10th grade) the source case was treated with antibiotics several times for "bronchitis." A chest radiograph was not obtained until February 1992, at which time cavitary disease was found. Bronchoscopy was performed and a specimen was smear positive for AFB; TB was diagnosed, and treatment with isoniazid, rifampin, and pyrazinamide was initiated. No sputum specimens were obtained before bronchoscopy. The isolate from bronchoscopy was sent to the county health department laboratory, which confirmed/]/L, taJaerculosL4 resistant to ISE and reported the finding to the local health departmertt. Neither the physician nor the hospital that processed the bronchoscopic specimen had reported the case to the health department. The private physician obtained a single follow-up sputum specimen in May 1992, which was smear and culture negative. A case report was initiated by t h e health department in duly 1992. Pharmacy record reviews indicated that prescriptions for antituberculous medications were filled inconsistently, and the
source case's adherence with treatment was not mentioned in the medical record at the private physician's office. In ,January 1995, bacteriologic moni~ toting of this patient's response to therapy was instituted at the request of the health department. Of 10 sputum specimens collected by the private phvsician between J a n u a r y 1, 1995, and ,July 51, 1995, seven were culture positive for ISF,-resistant/ILL, lubercu[oaiJ, and three were AFB smear positive. No contact investigation was initiated until May 1993 when other cases in the high school were recognized and transmission was suspected, The county health department assumed care of the source case with directly observed therapy in dune 1993, and the student completed a course of therapy. From the start of symptoms in January 199l until May 1993, the source case continued to attend school. STUDENT WITH TREATHENT FAILURE (CASE 2)
Case 2 was born in the United States, had no prior history of TB infection or disease, had a cough develop in January 1993, and was first diagnosed with TB in May 1993 while in i l t h grade. Sputum specimens were AFB smear positive, the isolate was ISE resistant, and chest radiograph showed bilateral upper lobe cavities. Isoniazid, rifampin, pyrazinamide, and ethambutol were started. Case 2 declined directly observed therapy by the local health department and was t r e a t e d by a managed care organization. The first of two sputum specimens obtained in August 1993 was smear and culture nega865
RIDZON ET AL.
THE JOURNAL OF PEDIATRICS DECEMBER 1997
(23%) had TSTs measuring ->10 mm induration (Table), and foreign-born students had significantly higher rates of pos1 ================================================================================================================================================================================ source case itive TSTs than U.S.-born students, 37% ============================== 211 versus 9%. Among U.S.-born students, 3 ~ compared with the 9th grade, the relative * Januaryto Juneonly risk of a positive TST was increased in the 4-** Sept~berto J~uaW 12th graders (relative risk = 6.1) and 1993 5 IIII ~ lnfsctlouspedod graduates (relative risk = 3.0). ..Q 0 Diagnosis o f TB 6 E School-wide screening was requested in ~ Expommto Source Case z spring 1994 for all previously negative 7 [ ~ Exposureto Case 2 students because of possible transmission 8-o from Case 2. Of the 928 students tested, 9-94 (10%) had a positive TST. The risk of 10 e a positive TST among students who were classroom contacts of Case 2 during the 11 1993-94 school year was increased com12 [ ] I ~, pared with students without classroom 13 I I I • contact to any case of TB (22 of 110 verF/g. 5. Exposure and infectious periods of students with TB. '*Januaryto june Only; **September to sus 54 of 616, relative risk 2.3, 95% CI 1.4 January Onl)~ Cross-hatched bars denote infectious periods; solid bars denote exposure to source case; to 3.8). Of the classroom contacts of Case open bars denote exposure to Case 2; circles denote dates when TB was diagnosed. 2, 20% had a positive TST. Of these, 21 of 22 (95%) had documented skin test contire; no smear was performed on the seccases were exposed during different classversions during the 1993-94 school year; ond specimen, but culture was positive for es. Case 12 was a classroom contact of the 14 of 21 (67%) converted their TST beM. tuberculasls. A single sputum specimen source case during 1990-91 and also of tween March 1994 and June 1994. obtained in September 1993 was smear Case 2 in spring 1993; pulmona W TB was O f the remaining 72 students with posand culture negative. On three of five ofdiagnosed in October, 1995 (Fig. 3). itive TSTs without classroom contact to Three remaining cases did not have Case 2, risk of a positive TST was not asflee visits between September 1, 1995, and December 31, 1995, Case 2 comclassroom contact with the source case. sociated with classroom contact with any other student with potentially infectious plained of cough and was treated with anCase 2 was potentially infectious from TB during the 1995-94 school year (one tibiotics for "bronchitis." The next sputum J a n u a W through M a y 1995 while in the l l t h grade and during the following outbreak-related case and two nonoutspecimen obtained in F e b r u a W 1994 school year while in the 12th grade. Case break-related cases). As in the fall screenshowed +4 AFB, was culture positive for ing, foreign-born students were signifi10 had classroom contact with Case 2 in ~L. tuberculosis, and was resistant to I S E cantly more likely to have a positive TST with newly acquired resistance to rispring 1995, had a documented skin test than U.S.-born students, 54 of 360 versus conversion at that time, and was diagfampin and ethambutol. Three months nosed with pleural TB in J u l y 1993. Case 18 old58, (relative risk 3.8, 95% CI 2.3 to later, a right upper lobectomy was per6.4). In the spring 1994 screening, 32 for8 hacl no classroom contact with either the formed to control extensive cavitaW diseign-born students with positive TSTs source case or Case 2, but was identified ease. Review of the outpatient medical were examined, and 28 (88%) had evias a friend of both. Case 15 was deterrecords showed that adherence to medicadence of a scar from Calmette-Gudrin mined to be an outbreak case based on tions from M a y 1993 to Februa W 1994 vaccination. drug susceptibility results and D N A finwas not addressed or documented. gerprint; however, no social or classroom PREVENTIVE THERAPY EPIDEMIOLOGIC INVESTIGATIONS OF contact with another case was identified. CASES Preventive therapy recommendations Contact investigations around all potenfor each person with a positive TST was tially infectious cases showed that only Contact investigations were performed one of three regimens. Those who rethe source case and Case 2 showed evifor all potentially infectious cases. The ceived isoniazid included the following: dence ofz~. tuberculosiJ transmission. source case was potentially infectious for (1) new entrants to the school, (2) those 29 months, and 9 of the 12 other cases (inTUBERCULOUS INFECTION IN HIGHwith a prior positive TST who had not recluding Case 2) had classroom exposure SCHOOL STUDENTS ceived preventive therapy, (3) those withto the source case during this period. Of out exposure to isoniazid-resistant cases. in September 1993, 94% of the students these, five were exposed during the latter Rifampin-preventive therapy was recomat the school and 158 of 347 (40%) memhalf of the 1990-91 school year in a formended for 157 students with exposure to bers of the already graduated class of 1995 eign language class that involved frequent an isoniazid-resistant case. For 22 classwere tested. Overall, 292 of 1265 students conversation and singing. The other four School year I 1990"91"1
866
I
School year 1991-92
I
I
School year 1992-93
I
School year 11993"94"~
RIDZON ET AL.
THE JOURNAL OF PEDIATRICS
Volume 13 I, Number 6 TaMe. Prevalence oftubecculous infection in high-school students, fall 1993
room contacts of Case 2 exposed during the 1995-94 school year while that case's isolate was resistant to five antituberculous drugs (including isoniazid and rifampin), pyrazinamide and ofloxaein preventive therapy was recommended. 20 Students receiw.~d directly observed preventive therapy at the school and were monitored closely for adverse effects by school nurses. Those who received isoniazid or rifampin tolerated the regimen with minimal side effects. The pyrazinamide and ofloxacin was stopped because of adverse effects in 64% of those who started the regimen. 21 Students who received preventive therapy were followed up for 2 years with symptom checks and chest radiographs. As of 1996, no cases of TB associated with this outbreak have been reported in any student.
DISCUSSION Two generations of transmission of drug-resistant TB among students in this California high school is supported by epidemiologic and laboratory evidence. All eight available isolates from the outbreak-related cases had identical D N A fingerprints. The results of the two screenings demonstrate that approximately g0% of the students in this school had a positive TST. The fall 19cl5 screening documented that, compared w{th other grades, relative risks of infection for the 12th graders and 1995 graduates were increased and mirrored the clustering of TB cases in these
two groups. Extensive transmission appears to have been from two cases, both of whom had cavita W disease, prolonged symptoms with cough, and multiple AFBpositive sputum smears. This transmission occurred predominantly in persons with close or prolonged contact with either case because all but one outbreak-related case had classroom or social contact with the source case and/or Case 2. During the spring 1994 screening, there were 21 documented TST conversions among classmates of Case 2. A large number of students with positive TSTs were not classroom contacts of either of the infectious cases. It is possible that infection may have been transmitted from one of the other potentially infectious students; however, the available data do not support this. One explanation for the high number of positive TSTs may be exposure outside the high school or exposure by casual contact in the school. In addition, a higher proportion of foreignborn students had positive skin tests than U.S.-born students, which may be due to old ~ tuJ~ercuJasia infection or cross-reactivity to prior bacille Calmette-Gutrin vaccination. 22 Numerous opportunities were missed to prevent and control this outbreak. These include delayed diagnosis and reporting of TB, incomplete evaluation by health care providers and the local health department, inappropriate monitoring of infectiousness, inadequate attention to adherence to treatment, and poor communication between providers and the health department. Both contagious cases had
multiple treatment for "bronchitis" without evaluation, despite the knowledge of ISE-resistant TB in the second case. In the management of all persons with pulmona W TB, infectiousness and response to therapy should be monitored. 23 For both cases, there was no documentation that three consecutive sputum specimens had been obtained by the private providers to confirm conversion of sputum to culture and smear negativity. Treatment with an inadequate drug regimen is the most common means by which resistant ~ ta/~erclt[oJia organisms are acquired, and persons with TB caused by drug-resistant organisms are much more likely to have treatment failure than those with TB caused by drug-susceptible isolates. 25 Ensuring compliance is essential to proper treatment of all cases of TB. The most successful way to accomplish this is through directly observed therapy, and directly observed therapy is strongly recommended for all persons with drug-resistant disease. 24 Neither the source case nor Case 2 had compliance with prescribed medicines closely monitored, and neither began treatment with D O T until serious complications developed. School-based screening for TB infection was once a widespread practice. However, in areas of low prevalence of infection, universal screening has been shown not to be cost-effectlve,26 and current recommendations suggest screening in high-risk populations only. 27 In a school such as the high school described here, where half the students are foreignborn from countries where TB is highly 867
RIDZON ET AL.
endemic, where evidence of transmission of TB among students exists and risk of exposure is high, screening may be considered an effective T B prevention a n d control tool. Screening should accompany monitoring for signs and symptoms of active TB and prompt treatment and followup of active cases. W i t h the combination of an undiagnosed case of drug-resistant TB, a setting such as a school, a population at increased risk of having a disease develop once infected, such as adolescents, 98 and nonadherence with TB prevention a n d treatment guidelines, a serious outbreak such as this one may be the result. In an adolescent, the diagnosis of TB should be considered with presentation of a persistent cough not improved with antibiotics, particularly if the person is from a count W with a high incidence of TB. With prompt diagnosis and appropriate treatment and monitoring of T B cases a n d timely a n d complete contact investigations, outbreaks such as this can be prevented. The patient management problems in the two students described offer a sobering picture of the difficulties that may be encountered as traditional public health services, such as TB clinics, are folded into managed care and private settings. Drug-resistant cases should be managed in consultation with an expert. Public health services have been delivered in categorical programs, each with its own expertise. These programs are at risk of being eradicated as health departments close or consolidate.
We thank Greg Andrews, Kenneth Castro, AID, Mark Fassell, Walter Ihle, and Brenda Mitchell from the Division of Tuberculosis Elimination, Centersfor Disease Control and Prevention, Atlanta, Georgia; Jan Curry, Kathy Higgins, Douglas Moore, PhD, Hildy Meyers, AID, Hugh Stallworth, AID, and Joe Vargas from Orange Counly Health Care Agency, Santa Ana, California; Jack Crawford, PhD from the Division of AIDS, STD, and TB Laboratory Research, Centers for Disease Control and Prevention, Atlanta, Georgia;Kate Cummings, Peggy Falknor, and Start Morita fi'om the California Department
868
THE JOURNAL OF PEDIATRICS DECEMBER 1997
of Health Services, Berkeley, California; and the high-schoolstudents and staff
REFERENCES 1. CDC. National action plan to combat multidrug-resistant tuberculosis. MMWR Morbid MortalWldy Rep 1992;41(RR-11):1-48. 2. Ecllin BR, Tokars dI, Grieco MH, et al. An outbreak of mulfidrug-resistant tuberculosis among hospitalized patients with the acquired immunodeficlency syndrome. N Engl J Med 1992;326:1514-21. 3. Pearson ML, Jereb JA, Frieden TR, et al. Nosocomial transmission of multidrug-resistant Mycobacteriamtuberculasia:a risk to patients and health care workers. Ann Intern Med 1992;t17:191-6. 4. Beck-Sague (3, Dooley SW,,Hutton MD, et al. Hospital outbreak of multldrug-resistant Alycobacteriumtubereu[osiainfections: factors in transmission to staff and HIV-infeeted patients. JAMA 1992;268:1280-6. 5. Valway SE, Richards SB, Kovacovich J, Greifinger RB, Crawford JT, Dooley SW. Outbreak of multi-drug-resistant tuberculosis in a New York State prison, 1991. Am J Epidemiol 1994;140:113-22. 6. CDC. Drug-resistant tuberculosis among the homeless-Boston. MMWR Morbid Mortal Wldy Rep 1985;34:429-31. 7. CDC. Outbreak of multidrug-resistant tuberculosis-Texas, California, and Pennsylvania. MMWR Morbid Mortal Wldy Rep I990;39:369-72. 8. CDC. Interstate outbreak of drug-resistant tuberculosis involving children ~ California, Montana, Nevada, Utah. MMWR Morbid Mortal Wldy Rep 1983;39:516-8. 9. Reves R, Blakey, D, Snider Jr D, Farer LS. Transmission of multiple drug-resistant tuberculosis: report of a school and community outbreak. Am d Epidemiul 1981; 113:423-35. 10. Sacks Jd, Brenner ER, Breeden DC, Anders HM, Parker RL. Epidemiolog¢ of a tuberculosis outbreak in a South Carolina junior high school. Am J Public Health 1985;75'.361-5. 11. Huge CW,, Fisher L, Donnell, Jr D, et al. Risk factors for transmission of ~lycobacterium tuberculosiain a primary school outbreak: lack of racial difference in susceptibility to infection. Am d Epidemiol 1994; 139:520-30. 12. Arlano E, Besozzi G, Belloni AM, et al. A school- and community-based outbreak of Mycobacteriumtuberca[osiJin Northern Italy, 1992-3. Epidemiol Infect 1994;113:83-93.
13. Lincoln EM. Epidemics of tuberculosis. Adv Tuberc Res 1965;14:157-201. 14. Hyge TV. Epidemic of tuberculosis in a state school, with an observation period of about 3 years. Acta Tuberculosis Stand
1947;21:1-57. 15. Core curriculum on tuberculosis. What the clinician should know. 3rd ed. Atlanta: Centers for Disease Control and Prevenfon; 1994. 16. Kent PT, Kubica G. Public health mycobacteri01ogy:a guide for the level III laboratory. US Department of Health and Human Services. Atlanta: Centers for Disease Control and Prevention; 1985. p. 159-84. 17. Van Embden JD, Cave MD, Crawford JT, et al. Strain identification of ~Iycobacteriam tuberculosisby DNA fingerprinting: recommendations for a standardized methodolo~¢. J Clin Microbiol 1993;51:4060. 18. Haas WH, Butler WR, Woodley CL, Crawford JT. N~ixeddinker polymerase chain reaction: a new rapid method for rapid fingerprinting of isolates of Mycobacteriumtubercu[oslscomplex. Clin Microhiol t993;31:1293-8. 19. Dean AG, Dean JA, Burton AH, et al. Epi Info, Version 5: a word processing database and statistics program for epidemiolo~y on micro-computers. Stone Mountain (GA): USD, Incorporated; 1990. 20. CDC. Management of persons exposed to multidrug-resistant tuberculosis MMWR Morbid Mortal WMy Rep 1992;41:(RR11)1-8. 21. Ridzon R, Meador J, Maxwell R, Higgins K, Weismuller E Onorato IM. Asymptomarie hepatitis in persons who received alternative preventive therapy with pyrazinamide and ofloxacin. Clin Infect Dis. 1997;24:1264-5. 22. Menzies R, Vissandjee B. Effect of bacille Calmette-Gudrin vaccination on tuberculin reactivity. Am Rev Respir Dis 1992; 145:621-5. 23. ATS/CDC. Treatment of tuberculosis and tuberculous infection in adults and children. Am J Respir Crit Care Med 1994; 149:135%74. 24. Iseman MD. Treatment of muhidrug-resistaut tuberculosis. N Engl J Med 1993; 329:784-91. 25. Mohle-Boetani JC, Miller B, Halpern M, et al. School-based screening for tuberculous infection. JAMA 1995;274:613-9. 26. American Academy of Pediatrics. Update on tuberculosis skin testing of children. Pediatrics 1996;97:282-3. 27. Starke JR, Jacobs RF, Jereb J. Resurgence of tuberculosis in children. J Pedlatr 1992;120:839-55.
Methods: To investigate potential transmission of drug-resistant tuberculosis, a retrospective cohort study with case investigation and screening by tuberculin skin tests and symptom checks was conducted in a high school of approximately 1400 students. Current and graduated high-school students were included in the investigation. D N A fingerprinting of available isolates was performed.
Results: Eighteen students with active tuberculosis were identified, Through epidemiologic and laboratory investigation, 13 cases were linked; 8 entered 12th grade in fall 1993; 9 of 13 had positive cultures for Alycobacterium tuberculosis with isoniazld, streptomycin, and ethionamide resistance, and all 8 available isolates had identical D N A fm:
In the United States drug-resistant tuberculosis has become a public health problem.1 Most reported outbreaks of drug-resistant TB have occurred in health care settings and have involved primalqly human immunodeficiency virus-infected patients. 2-5 Because of congregate conditions and extensive contact among students, schools are environments where transmission of TB occurs. 6"14 In this report we describe findings of a large outbreak investigation of drug-resistant TB in a California high school with approximately 1400 students.
gerprints. No staff member had tuberculosis. One student remained infectious for 29 months, from J a n u a r y 1991 to June 1993, and was the source case for the outbreak. Another student was infectious for 5 months before diagnosis in May 1993 and was a treatment failure in February 1994 with development of rifampin and ethambutol resistance in addition to isoniazid, streptomycin, and ethionamide. In the fall 1993 screening, 292 of 1263 (23%) students tested had a positive tuberculin skin test. Risk of infection was highest among 12th graders and classroom contacts of the two students with prolonged infectiousness. An additional 94 of 928 (10%) students tested in spring 1994
METHODS Epidemiologic Investigation
had a positive tuberculin skin test; 22 were classroom contacts of the student with
DEFINITIONS
treatment failure and 21 of these had documented tuberculin skin test conversions.
Case finding was conducted from the period of start of symptoms of the first outbreak case until the end of our investigation. A case of TB was defined as a current or former high-school student with signs and symptoms compatible with TB, diagnosis between January 1, 1991, and June 30, 1994, and laboratory evidence of TB or a positive tuberculin skin test. Laboratory evidence of TB included a clinical specimen that was smear positive for acid-fast bacilli, histologic evidence of TB, or culture positive for AlycobacterizLm ttaSerculasis. A case of multidrug-resistant TB was defined as a TB case with an isolate resistant to at least isoniazid and rifampin.
Conclusion: E~*ensive transmission of drug-resistant tuberculosis was documented in this high school, along with missed opportunities for prevention and control of this outbreak. Prompt identification of tuberculosis cases and timely interventions should help reduce this public health problem. ( J Pediatr 1997; 131:863-8)
From the Epidemic lntell~qeneeService, Epidemiology Program Office, Division of Tuberculosis E/imbmtion, Center, for Disease Controland Prevention, Atlanta, Georgia;Orange Count. Hea[rb CareAgency, Santa A~vz, California; California Department of Health Services, Berkeley;Dids hm of AIDS, STD, and TB Laboratmy Research, Centersfor Di~e,we Control and Prevention,Atlanta.
Submitted for publicationAug. 6, 1996; accepted Mar. 20, 1997. Reprint requests: Ren& Ridzon, MD, Divisionof TuberculosisElimination,Centers for Disease Control and Prevention, 1600 Clif}onRd., Atlanta,GA 30333. Copyright© 1997by Mosby-YearBook, Inc. 0022-3476/97/$5.00 ~ 0 9/21/82112
863
THE JOURNAL OF PEDIATRICS DECEMBER 1997
RIDZON ET AL.
Cases 6
•
Outbreak-relatedc a s e
[ ~ Nonoutbreak-related c a s e
Symptom Onset for Source case
1
Source Case Diagnosed with TB
N Jan- Apr- dulMar Jun Sep
OctDec
Jan- AprMar Jun
1991
Jul-
Oct-
Sep
Dec
1992
Jan- AprMar Jun
Jul-
Oct-
Sep
Dec
1993
Date of Diagnosis
F/g. L Casesof TB by date of diagnosis,California High School,
For persons from whom an isolate of 3/L
tuberculosis was obtained, an outbreak-related TB case was a case whose isolate demonstrated I S E resistance and had identical D N A fingerprints. If no isolate was obtained, the case was considered outbreak-related if an epidemiologic link, defined as a household contact, friend, or classroom contact of an outbreak-related case, was found. Cases were considered potentially infectious if evidence of pulmonary disease by chest radiograph was found or if a sputum or bronchoscopy specimen was either A F B smear-positive or culture-positive for M. tuberculosis. For this investigation, the beginning of an individual's infectious period was the date of cough onset or 4 weeks before the collection date of the first culture-positive specimen, whichever was earlier. It extended until the date when the first of three consecutive specimens, obtained on separate days, was smear and culture negative. If sputa were not obtained after start of therapy, the infectious period extended until 2 weeks after appropriate antituberculous medications were started. CASE FINDING AND SCHOOL-WIDE T S T SCREENING
To find cases of active TB and TB infection at the high school, we conducted a 864
school-wide TST screening in fall 1993. Students and staff with a documented prior positive TST were not rescreened. We conducted a second school-wide screening in spring 1994 because three students attending the school were diagnosed with TB as a result of the fall screening and one student first diagnosed in M a y 1993 failed treatment and had A F B smear-positive sputum in February 1994. Each person screened was interviewed for symptoms consistent with TB and completed a self-administered questionnaire. TST screening was performed by the M a n t o u x method using 5 TU Tubersol (brand names used for identification purposes only) purified protein derivative. TSTs were applied and read 48 to 72 hours later by county health department nurses. For the purpose of providing preventive therapy for contacts of infectious cases, a positive TST was defined as >5 mm induration. Because of the large proportion (50%) of students horn in countries of high TB incidence (77% and 9% of foreign-born students were from Vietnam and Mexico, respectively) and potential positive reactions caused by bacille Calmette-Gu4rin vaccination, to increase specificity, we defined a TST of >10 mm as positive in the epidemiologic study. A TST conversion was defined as
an increase in induration of at least 10 mm in the previous 2 years. 15 All students with positive TSTs had chest radiographs perforraed. To identify classroom contacts of infectious cases, class schedules were obtained. A cohort of 32 foreign-born students was examined for the presence of a Bacille Calmette-Gu4rin scar. Additional cases of TB were Sought by matching the state and county TB registries from January 1990 through June 1994 to the school roster that contained names of staff, current students, "transfer students, and students who graduated in 1992 and 1993. County TB clinic records were reviewed for persons 13 to 20 years old, and county public health laboratoW records were reviewed to identify persons whose isolate had a susceptibility pattern that included I S E resistance. All available medical records for persons identified by these methods were reviewed.
Laboratory Investigation Susceptibility testing of ~L. tuberculosis isolates was conducted using the proportional method.16 Available isolates were sent to the Centers for Disease Control and Prevention for D N A fingerprinting using IS6110 restriction fragment length polymorphism and mixed-linker polymerase chain reaction methods. 17'18
Statistical Analysis Data were analyzed using Epi Info, version 5.0.19 For analysis of the fall 1993 TST screening data, the 9th grade (new high-school entrants not likely to have been exposed to outbreak-related cases) was used as a referent group to estimate grade-specific relative risks for a positive TST. Students in grades 10 through 12 who were new transfers into the school and therefore not exposed during the prior year were excluded from analysis of f~l screening results. Analyses were also stratified by birthplace to eliminate confounding from infections acquired outside the United States.
RESULTS During spring of 1993, two cases of active pulmonary TB were identified among 1 lth grade students. Through contact in-
~IDZON ET AL.
THE JOURNAL OF PEDIATRICS
Volume 131, Number
6
vestigations conducted by the local health department, three additional TB cases were identified in 1 lth grade students. ,4£. tuberculosi~ isolates from four of the five cases had identical susceptibility patterns with resistance to ISE. An outbreak investigation was initiated in fall 1993. ACTIVE TUBERCULOSIS IN HIGHSCHOOL STUDENTS
Cough Onset
1 I ~v~
A total of 18 TB cases were identified among current high-school students and 1992 and 1995 graduates. Cases were diagnosed from January 1991 to November 1995(Fig. 1). There were no cases among staff. Five cases were not related to the outbreak. Four were culture positive for z//L. tuberculos/a, all had drug susceptibility patterns that did not include ISE resistance, and each had a distinct D N A fingerprint. The fifth case was diagnosed on clinical criteria and did not have an epidemiologic link to any outbreak-related case. On the basis of epidemiologic and laboratory data, 13 cases were outbreak-related. Ten cases were female, 10 were Asian, and 9 were foreign born. Four cases graduated in spring 1993, eight were in 12th grade, and one was in 10th grade. Nine of these 15 cases were culture-positive; eight had d/Z. tubereu[osia isolates that were initially resistant to ISE, and one was initially resistant to ISE and ethambutol. All eight isolates available for D N A fingerprinting (including the ISE/ethambutolresistant isolate) had identical patterns. In addition to the nine culture-positive cases, one other case demonstrated AFB on biopsy specimen examination of lymphoid tissue, one had granuloma on pleural biopsy specimen, and the remaining two were diagnosed clinically. Nine had pulmonary disease, three had pleural disease, and one had lymphadenitis. Seven cases attended school[ while they had pulmonary disease. Of these seven, three had multiple AFB-srnear positive sputa, t h r e e had multiple sputum smears negative for AFB, one had a single sputum specimen with rare AFB and multiple other AFB smear-negative specimens. On the basis of medical record review and clinical course and analysis of screening data, 2 of these 1S cases were determined to be responsible for transmission
Lab report to HD
TB diagnosis
J
t 1
.......... - ........
County Sputum contacts County culture physician assumes negative
J cae J i
T Tr lllll
I/R/Z started
EMB added
Positive sputum cultures
- Ethmb~toI
Jan
Apr
Jul
Oct
Jan
Apr
I
I
I
I
I
I
1991
Jui
1992
Oct
Jan 1993
Apr
Jul
Oct
Jan 1994
F ~ . 2. Sourcecasehistory Solid bar representsthe case'sinfectiousperiod,HD, Health Department; I/R/Z, isoniazidrifampin,and pyrazinamide;EJNB,ethambutol,
during this outbreak. Their clinical courses are summarized below. SOURCE CASE
The source case entered 12th grade in fall 1993. The student was born in Southeast Asia, moved to the United States as a child, and by history had a positive TST and was treated with a 6-month course of isoniazid. Cough developed in J a n u a r y 1991 and persisted through August 1993 (Fig. 2). From Oanua W 1991 until February 1992 (while in 9th and 10th grade) the source case was treated with antibiotics several times for "bronchitis." A chest radiograph was not obtained until February 1992, at which time cavitary disease was found. Bronchoscopy was performed and a specimen was smear positive for AFB; TB was diagnosed, and treatment with isoniazid, rifampin, and pyrazinamide was initiated. No sputum specimens were obtained before bronchoscopy. The isolate from bronchoscopy was sent to the county health department laboratory, which confirmed/]/L, taJaerculosL4 resistant to ISE and reported the finding to the local health departmertt. Neither the physician nor the hospital that processed the bronchoscopic specimen had reported the case to the health department. The private physician obtained a single follow-up sputum specimen in May 1992, which was smear and culture negative. A case report was initiated by t h e health department in duly 1992. Pharmacy record reviews indicated that prescriptions for antituberculous medications were filled inconsistently, and the
source case's adherence with treatment was not mentioned in the medical record at the private physician's office. In ,January 1995, bacteriologic moni~ toting of this patient's response to therapy was instituted at the request of the health department. Of 10 sputum specimens collected by the private phvsician between J a n u a r y 1, 1995, and ,July 51, 1995, seven were culture positive for ISF,-resistant/ILL, lubercu[oaiJ, and three were AFB smear positive. No contact investigation was initiated until May 1993 when other cases in the high school were recognized and transmission was suspected, The county health department assumed care of the source case with directly observed therapy in dune 1993, and the student completed a course of therapy. From the start of symptoms in January 199l until May 1993, the source case continued to attend school. STUDENT WITH TREATHENT FAILURE (CASE 2)
Case 2 was born in the United States, had no prior history of TB infection or disease, had a cough develop in January 1993, and was first diagnosed with TB in May 1993 while in i l t h grade. Sputum specimens were AFB smear positive, the isolate was ISE resistant, and chest radiograph showed bilateral upper lobe cavities. Isoniazid, rifampin, pyrazinamide, and ethambutol were started. Case 2 declined directly observed therapy by the local health department and was t r e a t e d by a managed care organization. The first of two sputum specimens obtained in August 1993 was smear and culture nega865
RIDZON ET AL.
THE JOURNAL OF PEDIATRICS DECEMBER 1997
(23%) had TSTs measuring ->10 mm induration (Table), and foreign-born students had significantly higher rates of pos1 ================================================================================================================================================================================ source case itive TSTs than U.S.-born students, 37% ============================== 211 versus 9%. Among U.S.-born students, 3 ~ compared with the 9th grade, the relative * Januaryto Juneonly risk of a positive TST was increased in the 4-** Sept~berto J~uaW 12th graders (relative risk = 6.1) and 1993 5 IIII ~ lnfsctlouspedod graduates (relative risk = 3.0). ..Q 0 Diagnosis o f TB 6 E School-wide screening was requested in ~ Expommto Source Case z spring 1994 for all previously negative 7 [ ~ Exposureto Case 2 students because of possible transmission 8-o from Case 2. Of the 928 students tested, 9-94 (10%) had a positive TST. The risk of 10 e a positive TST among students who were classroom contacts of Case 2 during the 11 1993-94 school year was increased com12 [ ] I ~, pared with students without classroom 13 I I I • contact to any case of TB (22 of 110 verF/g. 5. Exposure and infectious periods of students with TB. '*Januaryto june Only; **September to sus 54 of 616, relative risk 2.3, 95% CI 1.4 January Onl)~ Cross-hatched bars denote infectious periods; solid bars denote exposure to source case; to 3.8). Of the classroom contacts of Case open bars denote exposure to Case 2; circles denote dates when TB was diagnosed. 2, 20% had a positive TST. Of these, 21 of 22 (95%) had documented skin test contire; no smear was performed on the seccases were exposed during different classversions during the 1993-94 school year; ond specimen, but culture was positive for es. Case 12 was a classroom contact of the 14 of 21 (67%) converted their TST beM. tuberculasls. A single sputum specimen source case during 1990-91 and also of tween March 1994 and June 1994. obtained in September 1993 was smear Case 2 in spring 1993; pulmona W TB was O f the remaining 72 students with posand culture negative. On three of five ofdiagnosed in October, 1995 (Fig. 3). itive TSTs without classroom contact to Three remaining cases did not have Case 2, risk of a positive TST was not asflee visits between September 1, 1995, and December 31, 1995, Case 2 comclassroom contact with the source case. sociated with classroom contact with any other student with potentially infectious plained of cough and was treated with anCase 2 was potentially infectious from TB during the 1995-94 school year (one tibiotics for "bronchitis." The next sputum J a n u a W through M a y 1995 while in the l l t h grade and during the following outbreak-related case and two nonoutspecimen obtained in F e b r u a W 1994 school year while in the 12th grade. Case break-related cases). As in the fall screenshowed +4 AFB, was culture positive for ing, foreign-born students were signifi10 had classroom contact with Case 2 in ~L. tuberculosis, and was resistant to I S E cantly more likely to have a positive TST with newly acquired resistance to rispring 1995, had a documented skin test than U.S.-born students, 54 of 360 versus conversion at that time, and was diagfampin and ethambutol. Three months nosed with pleural TB in J u l y 1993. Case 18 old58, (relative risk 3.8, 95% CI 2.3 to later, a right upper lobectomy was per6.4). In the spring 1994 screening, 32 for8 hacl no classroom contact with either the formed to control extensive cavitaW diseign-born students with positive TSTs source case or Case 2, but was identified ease. Review of the outpatient medical were examined, and 28 (88%) had evias a friend of both. Case 15 was deterrecords showed that adherence to medicadence of a scar from Calmette-Gudrin mined to be an outbreak case based on tions from M a y 1993 to Februa W 1994 vaccination. drug susceptibility results and D N A finwas not addressed or documented. gerprint; however, no social or classroom PREVENTIVE THERAPY EPIDEMIOLOGIC INVESTIGATIONS OF contact with another case was identified. CASES Preventive therapy recommendations Contact investigations around all potenfor each person with a positive TST was tially infectious cases showed that only Contact investigations were performed one of three regimens. Those who rethe source case and Case 2 showed evifor all potentially infectious cases. The ceived isoniazid included the following: dence ofz~. tuberculosiJ transmission. source case was potentially infectious for (1) new entrants to the school, (2) those 29 months, and 9 of the 12 other cases (inTUBERCULOUS INFECTION IN HIGHwith a prior positive TST who had not recluding Case 2) had classroom exposure SCHOOL STUDENTS ceived preventive therapy, (3) those withto the source case during this period. Of out exposure to isoniazid-resistant cases. in September 1993, 94% of the students these, five were exposed during the latter Rifampin-preventive therapy was recomat the school and 158 of 347 (40%) memhalf of the 1990-91 school year in a formended for 157 students with exposure to bers of the already graduated class of 1995 eign language class that involved frequent an isoniazid-resistant case. For 22 classwere tested. Overall, 292 of 1265 students conversation and singing. The other four School year I 1990"91"1
866
I
School year 1991-92
I
I
School year 1992-93
I
School year 11993"94"~
RIDZON ET AL.
THE JOURNAL OF PEDIATRICS
Volume 13 I, Number 6 TaMe. Prevalence oftubecculous infection in high-school students, fall 1993
room contacts of Case 2 exposed during the 1995-94 school year while that case's isolate was resistant to five antituberculous drugs (including isoniazid and rifampin), pyrazinamide and ofloxaein preventive therapy was recommended. 20 Students receiw.~d directly observed preventive therapy at the school and were monitored closely for adverse effects by school nurses. Those who received isoniazid or rifampin tolerated the regimen with minimal side effects. The pyrazinamide and ofloxacin was stopped because of adverse effects in 64% of those who started the regimen. 21 Students who received preventive therapy were followed up for 2 years with symptom checks and chest radiographs. As of 1996, no cases of TB associated with this outbreak have been reported in any student.
DISCUSSION Two generations of transmission of drug-resistant TB among students in this California high school is supported by epidemiologic and laboratory evidence. All eight available isolates from the outbreak-related cases had identical D N A fingerprints. The results of the two screenings demonstrate that approximately g0% of the students in this school had a positive TST. The fall 19cl5 screening documented that, compared w{th other grades, relative risks of infection for the 12th graders and 1995 graduates were increased and mirrored the clustering of TB cases in these
two groups. Extensive transmission appears to have been from two cases, both of whom had cavita W disease, prolonged symptoms with cough, and multiple AFBpositive sputum smears. This transmission occurred predominantly in persons with close or prolonged contact with either case because all but one outbreak-related case had classroom or social contact with the source case and/or Case 2. During the spring 1994 screening, there were 21 documented TST conversions among classmates of Case 2. A large number of students with positive TSTs were not classroom contacts of either of the infectious cases. It is possible that infection may have been transmitted from one of the other potentially infectious students; however, the available data do not support this. One explanation for the high number of positive TSTs may be exposure outside the high school or exposure by casual contact in the school. In addition, a higher proportion of foreignborn students had positive skin tests than U.S.-born students, which may be due to old ~ tuJ~ercuJasia infection or cross-reactivity to prior bacille Calmette-Gutrin vaccination. 22 Numerous opportunities were missed to prevent and control this outbreak. These include delayed diagnosis and reporting of TB, incomplete evaluation by health care providers and the local health department, inappropriate monitoring of infectiousness, inadequate attention to adherence to treatment, and poor communication between providers and the health department. Both contagious cases had
multiple treatment for "bronchitis" without evaluation, despite the knowledge of ISE-resistant TB in the second case. In the management of all persons with pulmona W TB, infectiousness and response to therapy should be monitored. 23 For both cases, there was no documentation that three consecutive sputum specimens had been obtained by the private providers to confirm conversion of sputum to culture and smear negativity. Treatment with an inadequate drug regimen is the most common means by which resistant ~ ta/~erclt[oJia organisms are acquired, and persons with TB caused by drug-resistant organisms are much more likely to have treatment failure than those with TB caused by drug-susceptible isolates. 25 Ensuring compliance is essential to proper treatment of all cases of TB. The most successful way to accomplish this is through directly observed therapy, and directly observed therapy is strongly recommended for all persons with drug-resistant disease. 24 Neither the source case nor Case 2 had compliance with prescribed medicines closely monitored, and neither began treatment with D O T until serious complications developed. School-based screening for TB infection was once a widespread practice. However, in areas of low prevalence of infection, universal screening has been shown not to be cost-effectlve,26 and current recommendations suggest screening in high-risk populations only. 27 In a school such as the high school described here, where half the students are foreignborn from countries where TB is highly 867
RIDZON ET AL.
endemic, where evidence of transmission of TB among students exists and risk of exposure is high, screening may be considered an effective T B prevention a n d control tool. Screening should accompany monitoring for signs and symptoms of active TB and prompt treatment and followup of active cases. W i t h the combination of an undiagnosed case of drug-resistant TB, a setting such as a school, a population at increased risk of having a disease develop once infected, such as adolescents, 98 and nonadherence with TB prevention a n d treatment guidelines, a serious outbreak such as this one may be the result. In an adolescent, the diagnosis of TB should be considered with presentation of a persistent cough not improved with antibiotics, particularly if the person is from a count W with a high incidence of TB. With prompt diagnosis and appropriate treatment and monitoring of T B cases a n d timely a n d complete contact investigations, outbreaks such as this can be prevented. The patient management problems in the two students described offer a sobering picture of the difficulties that may be encountered as traditional public health services, such as TB clinics, are folded into managed care and private settings. Drug-resistant cases should be managed in consultation with an expert. Public health services have been delivered in categorical programs, each with its own expertise. These programs are at risk of being eradicated as health departments close or consolidate.
We thank Greg Andrews, Kenneth Castro, AID, Mark Fassell, Walter Ihle, and Brenda Mitchell from the Division of Tuberculosis Elimination, Centersfor Disease Control and Prevention, Atlanta, Georgia; Jan Curry, Kathy Higgins, Douglas Moore, PhD, Hildy Meyers, AID, Hugh Stallworth, AID, and Joe Vargas from Orange Counly Health Care Agency, Santa Ana, California; Jack Crawford, PhD from the Division of AIDS, STD, and TB Laboratory Research, Centers for Disease Control and Prevention, Atlanta, Georgia;Kate Cummings, Peggy Falknor, and Start Morita fi'om the California Department
868
THE JOURNAL OF PEDIATRICS DECEMBER 1997
of Health Services, Berkeley, California; and the high-schoolstudents and staff
REFERENCES 1. CDC. National action plan to combat multidrug-resistant tuberculosis. MMWR Morbid MortalWldy Rep 1992;41(RR-11):1-48. 2. Ecllin BR, Tokars dI, Grieco MH, et al. An outbreak of mulfidrug-resistant tuberculosis among hospitalized patients with the acquired immunodeficlency syndrome. N Engl J Med 1992;326:1514-21. 3. Pearson ML, Jereb JA, Frieden TR, et al. Nosocomial transmission of multidrug-resistant Mycobacteriamtuberculasia:a risk to patients and health care workers. Ann Intern Med 1992;t17:191-6. 4. Beck-Sague (3, Dooley SW,,Hutton MD, et al. Hospital outbreak of multldrug-resistant Alycobacteriumtubereu[osiainfections: factors in transmission to staff and HIV-infeeted patients. JAMA 1992;268:1280-6. 5. Valway SE, Richards SB, Kovacovich J, Greifinger RB, Crawford JT, Dooley SW. Outbreak of multi-drug-resistant tuberculosis in a New York State prison, 1991. Am J Epidemiol 1994;140:113-22. 6. CDC. Drug-resistant tuberculosis among the homeless-Boston. MMWR Morbid Mortal Wldy Rep 1985;34:429-31. 7. CDC. Outbreak of multidrug-resistant tuberculosis-Texas, California, and Pennsylvania. MMWR Morbid Mortal Wldy Rep I990;39:369-72. 8. CDC. Interstate outbreak of drug-resistant tuberculosis involving children ~ California, Montana, Nevada, Utah. MMWR Morbid Mortal Wldy Rep 1983;39:516-8. 9. Reves R, Blakey, D, Snider Jr D, Farer LS. Transmission of multiple drug-resistant tuberculosis: report of a school and community outbreak. Am d Epidemiul 1981; 113:423-35. 10. Sacks Jd, Brenner ER, Breeden DC, Anders HM, Parker RL. Epidemiolog¢ of a tuberculosis outbreak in a South Carolina junior high school. Am J Public Health 1985;75'.361-5. 11. Huge CW,, Fisher L, Donnell, Jr D, et al. Risk factors for transmission of ~lycobacterium tuberculosiain a primary school outbreak: lack of racial difference in susceptibility to infection. Am d Epidemiol 1994; 139:520-30. 12. Arlano E, Besozzi G, Belloni AM, et al. A school- and community-based outbreak of Mycobacteriumtuberca[osiJin Northern Italy, 1992-3. Epidemiol Infect 1994;113:83-93.
13. Lincoln EM. Epidemics of tuberculosis. Adv Tuberc Res 1965;14:157-201. 14. Hyge TV. Epidemic of tuberculosis in a state school, with an observation period of about 3 years. Acta Tuberculosis Stand
1947;21:1-57. 15. Core curriculum on tuberculosis. What the clinician should know. 3rd ed. Atlanta: Centers for Disease Control and Prevenfon; 1994. 16. Kent PT, Kubica G. Public health mycobacteri01ogy:a guide for the level III laboratory. US Department of Health and Human Services. Atlanta: Centers for Disease Control and Prevention; 1985. p. 159-84. 17. Van Embden JD, Cave MD, Crawford JT, et al. Strain identification of ~Iycobacteriam tuberculosisby DNA fingerprinting: recommendations for a standardized methodolo~¢. J Clin Microbiol 1993;51:4060. 18. Haas WH, Butler WR, Woodley CL, Crawford JT. N~ixeddinker polymerase chain reaction: a new rapid method for rapid fingerprinting of isolates of Mycobacteriumtubercu[oslscomplex. Clin Microhiol t993;31:1293-8. 19. Dean AG, Dean JA, Burton AH, et al. Epi Info, Version 5: a word processing database and statistics program for epidemiolo~y on micro-computers. Stone Mountain (GA): USD, Incorporated; 1990. 20. CDC. Management of persons exposed to multidrug-resistant tuberculosis MMWR Morbid Mortal WMy Rep 1992;41:(RR11)1-8. 21. Ridzon R, Meador J, Maxwell R, Higgins K, Weismuller E Onorato IM. Asymptomarie hepatitis in persons who received alternative preventive therapy with pyrazinamide and ofloxacin. Clin Infect Dis. 1997;24:1264-5. 22. Menzies R, Vissandjee B. Effect of bacille Calmette-Gudrin vaccination on tuberculin reactivity. Am Rev Respir Dis 1992; 145:621-5. 23. ATS/CDC. Treatment of tuberculosis and tuberculous infection in adults and children. Am J Respir Crit Care Med 1994; 149:135%74. 24. Iseman MD. Treatment of muhidrug-resistaut tuberculosis. N Engl J Med 1993; 329:784-91. 25. Mohle-Boetani JC, Miller B, Halpern M, et al. School-based screening for tuberculous infection. JAMA 1995;274:613-9. 26. American Academy of Pediatrics. Update on tuberculosis skin testing of children. Pediatrics 1996;97:282-3. 27. Starke JR, Jacobs RF, Jereb J. Resurgence of tuberculosis in children. J Pedlatr 1992;120:839-55.