mycobacterium tuberculosis transmission among high school

7/27/2019 Mycobacterium Tuberculosis Transmission Among High School

1/6

Pediatrics International(2005) 47, 180184

Original Article

Mycobacterium tuberculosis transmission among high school

students in Greece

CHRISTOS HADJICHRISTODOULOU,1 ANTONIS VASILOGIANNAKOPOULOS,

2

GEORGIA SPALA,

2

IRINI MAVROU,

2

VIRGINIA KOLONIA,

2

EVANGELOS MARINIS,

4

VASILIKI SYRIOPOULOU

3

AND MARIA THEODORIDOU

3

1

Medical School of the University of Thesalia, Department of Hygiene and Epidemiology, Larisa,2

National Centers for Surveillance and Intervention (NCSI), Athens, 3

First Pediatric Clinic of the

University of Athens Agia Sofia Hospital, Athens, and4

National Reference Laboratories for

Mycobacterium, Sotiria Hospital Athens, Greece

Abstract Background

: The aim of this study was to investigate the requirements and practical steps for screening of

Mycobacterium tuberculosis

(MTB) transmission among high school student populations in two regional high

schools of central Greece. Case-matched control populations from other regional schools were included.

Methods

: Case study of two indexed cases, 61 close contacts, 212 casual contacts and 369 controls wereinvestigated. Detailed questionnaires, tuberculin-skin test (PPD test), chest radiography, medical evaluation

and DNA fingerprinting of sputum isolates were used.

Results

: In case A, three (1.97%) of 152 close and casual contacts developed tuberculosis, and a further 25

(16.4%) were classified as infected. In contrast, none of the 121 close or casual contacts investigated for Case

B developed tuberculosis or were classified as infected. None of the control populations contained infected

individuals. Contacts of case A had a much higher risk (3.08 < RR = 22.29 < 161.69, P

< 0.001) of being

infected than contacts of case B. Two different strains of MTB were found responsible for these outbreaks.

Conclusion

: There was a considerable difference in the infectivity of the two cases presumably due to

environmental and clinical factors, although two different MTB strains were responsible. It is proposed that

the extent of case investigation should be individualized with particular emphasis placed among close

contacts.

Key words case investigation,Mycobacterium tuberculosis

, tuberculosis.

Morbidity and mortality as a result of tuberculosis has been

declining since the beginning of the twentieth century. Improve-

ments in diet, housing, living conditions and the advent of

effective chemotherapy contributed significantly to this decline.

However,Mycobacterium tuberculosis

(MTB) has reemerged

in the industrialized world over the last two decades.

1

The notification rate in Greece is about 910 new cases

per 100 000 per year, but there is a significant level of under-

reporting. From 1950 until 1988, Bacille Calmette-Guerinvaccination was routinely performed in the military service

and was recommended for children

12 years old. Since

1988 it has been recommended for all children

6 years

old, with a current estimated BCG coverage among children

of 30%.

At schools, there are special circumstances of crowding

and a particular age-specific susceptibility which favor the

emergence of tuberculosis microepidemics.

2

Two cases of

pulmonary tuberculosis in two different public high schools

of an urban region in central Greece were reported to the

National Center for Surveillance and Intervention (NCSI)

between December 1997 and April 1998.Case A, a 17-year-old Greek male student, resident of an

urban area of low socioeconomic status, was symptomatic

with fever and productive cough and was empirically treated

for pneumonia for more than 20 days without improvement.

At that point, chest X-ray revealed a right upper lobe cavitary

lesion. Sputum specimens were negative for acid-fast bacilli

but he was hospitalized and given isoniazid, rifampin, pyrazi-

namide and ethambutol for tuberculosis.

3

MTB was detected

later with the sputum culture.

Correspondence: Hadjichristodoulou Christos, Department ofHygiene and Epidemiology, University of Thesalia, Papakiriazi 22,41222, Larisa, Greece. Email: [email protected]

Received 19 February 2004; revised and accepted 30 August2004.


2/6

Tuberculosis transmission in high schools 181

Case B, a 15-year-old Greek female student, resident of

the same area as case A, was symptomatic with fever, cough

and hemoptysis for 4 days when she was hospitalized. Chest

X-ray revealed small cavitary lesions in the left upper

pulmonary lobe. Sputum specimens were positive for acid-

fast bacilli and she was administered isoniazid, rifampin,

pyrazinamid and ethambutol. The source of infection was

presumably her uncle who had been diagnosed with smear

positive pulmonary tuberculosis 1 year before her illness.

A tuberculosis case investigation team was set up to

investigate the extent of transmission in the family members,

schoolmates and teachers of these two cases compared to

controls (no contacts) in each case.

Methods

The epidemiological investigation included tuberculin skin

test screening (PPD test), completion of special question-

naires, medical evaluation and chest radiograph of individuals

with positive skin tests, and DNA fingerprinting of MTB

isolates. Control groups were included in the investigation tocompare the positive results in the PPD test among the BCG

vaccinated individuals, avoiding misinterpretation of the

results. The PPD test was done using the Mantoux method

with 5 IU PPD (Institute Pasteur-Merieux, Athens, Greece);

the test was performed and read by NCSI physicians and

health care workers. A positive PPD test was considered

when the induration was

10 mm. Given that some students

were BCG vaccinated, we considered as infected all who had

a Mantoux induration >15 mm and were vaccinated within

5 years. In addition, we considered as infected all who had

Mantoux induration >10 mm and were either non-vaccinated

or vaccinated more than 5 years previously.

4

Close contactsand casual contacts of the index cases were included in the

investigation by using the PPD test and a detailed question-

naire. Close contacts were defined as all students who

attended the same classroom (classroom mates) as the index

cases, and all family members. All other students in the same

class, but from different classrooms (classmates), and the

teachers of the school were considered as casual contacts.

Students and teachers from other schools with no contact

with the two cases were considered as controls. The controls

were randomly selected amongst students and teachers from

different schools of the same area, of the same age range, and

BCG vaccination status.

All infected individuals had chest radiography performed,

medical evaluation and received chemoprophylaxis or chemo-

therapy. To exclude other sources of infection, members

of the extended family (aunts, uncles and grandparents) of

infected individuals were screened with a tuberculin skin test

and a questionnaire was used to obtain information regarding

possible sources of infection. In addition, in some members

of the extended family chest X-ray was performed (on

clinical grounds).

The strains of MTB isolated from the two cases were sent

to the National Reference Laboratory for Mycobacterium in

Athens for susceptibility testing and then to Raymond Poincare

Hospital in France for DNA fingerprinting using 15611S

restriction fragment length polymorphism and mixed-linked

polymerase chain reaction methods.

Results

Tuberculin screening

Case A

A total of 183 contacts of case A were identified including

three family members, 149 students and 31 teachers. A total

of 229 students and teachers who had no contact with case A

were used as controls. Screening results were obtained from

118 (79.2%) of 149 students, three (100%) family members,

31 (100%) school teachers and 229 (100%) controls. The

PPD test was positive in 37 (24.4%) out of 152 casual

contacts and close contacts, comparing to 10 (4.4%) out of229 controls; all contacts had a higher risk of having a

positive PPD test (1.63 < RR = 3.14 < 6.06, P

< 0.001). Among

the close contacts, 15 (51.7%) out of 29 classroom mates

were PPD positive and no family members; among casual

contacts 19 (21.3%) out of 89 classmates, and three (9.7%)

out of 31 teachers were positive. The close contacts were at

a much higher risk (5.88 < RR = 11.84 < 23.87, P

< 0.01) of

having a positive tuberculin test compared to casual contacts

(Table 1).

Table 1

Case A tuberculin test results

Positive/Total Relative risk

P

-value

Close contacts Family 0/3 (0%) Classroom mates 15/29 (51.7%) 5.88 < 11.84 < 23.87 < 0.001

Casual contacts Teachers 3/31 (9.7%) 0.64 < 2.22 < 7.62 0.19Classmates 19/89 (21.3%) 2.37 < 4.89 < 10.10 < 0.001

Total 37/152 (24.4%) 1.63 < 3.14 < 6.06 < 0.001No contact (control) 10/229 (4.4%)


3/6

182 C Hadjichristodoulou et al

.

Case B

A total of 157 contacts of case B were identified including

four family members and 153 students; 140 students who had

no contact with case B were used as controls. Screening

results were obtained from 117 (74.5%) students and four

(100%) family members; no school teacher co-operated in

performing the PPD test. The PPD test was positive in 15

(12.4%) out of 121 casual contacts and close contacts

compared to nine (6.4%) out of 140 controls, indicating that

contacts were not at a greater transmission risk than the

control group (

P =

0.14; Table 2).

Infected individuals

Using the criteria mentioned above to define the infected

individuals, 14 (43.7%) close contacts and 11 (9.1%) casual

contacts of case A were classified as infected. Close contacts

had a higher risk of infection (2.00 < RR = 3.91 < 7.63,

P

< 0.001) in comparison with casual contacts. In total, 25

(16.4%) contacts (close and casual contacts) were classified

as infected showing a much higher risk (6.68 < RR = 44.73

< 355, P

< 0.0001) of being infected compared to the control

group.

In case B, none of the 121 contacts (close and casual

contacts) or 140 controls were classified as infected. Allcontacts in case B did not have a higher risk of being infected

than the control group.

All contacts (close and casual contacts) in case A had a

much higher risk of being infected (2.74 < RR = 19.90 < 144.7,

P

< 0.001) when compared to all contacts of case B (Table 3).

A total of 131 additional friends and members of the

extended family (aunts, uncles and grandparents) of the infected

cases were screened using the PPD test, in order to detect

other possible sources of infection. Therefore, a special

questionnaire was used and 20 individuals were examined

with chest X-ray. No case of tuberculosis was detected.

DNA fingerprinting analysis of the two isolates showed

two different strains susceptible to the usual preventive

chemotherapy.

Secondary tuberculosis cases

Case A

From December 1997 through to March 1998, three cases of

tuberculosis were diagnosed on clinical and laboratory

grounds among classroom mates of case A.

Case B

From March to July 1998, no cases of tuberculosis were

identified (Table 3). Despite the fact that there was a

difference in the risk of developing secondary tuberculosis

between the two index cases, this was not proved to be

statistically significant (0.25 < RR = 2.39 < 22.7, P

= 0.25).

Discussion

Schools are settings with high concentrations of young peoplewith little exposure to MTB

2,5

but conditions of crowding and

age-related susceptibility favor the emergence of tuberculosis

outbreaks.

6

It is obvious that there is a relation between the

hours spent together with the index case and the risk of

infectivity.

7

Emphasis is given in the need for identification

and prompt chemoprophylaxis/chemotherapy, especially in

vulnerable adolescent populations.

8

Primary care physicians

and public health authorities must assume a leading role in

the early detection and treatment of patients.

912

Table 2

Case B tuberculin test results

Positive/Total Relative risk

P

-value

Close contacts Family 0/4 (0%) Classroom mates 3/25 (12.0%) 0.54 < 1.87 < 6.42 0.39

Casual contacts Teachers Classmates 12/92 (13.1%) 0.89 < 2.03 < 4.62 0.13

Total 15/121 (12.4%) 0.88 < 1.93 < 4.25 0.14No contact (control) 9/140 (6.4%)

Table 3

Tuberculosis and infection among all contacts

Case A Case B Relative risk

P

-value

Tuberculosis 3/152 (1.2%) 0/121 (0.0%) 0.25 < 2.39 < 22.67 0.25Infection 25/152 (16.4%) 0/121 (0.0%) 2.74 < 19.90 < 144.7 < 0.001TB and infection 28/152 (18.4%) 0/121 (0.0%) 3.08 < 22.29 < 161.49 < 0.001


4/6

Tuberculosis transmission in high schools 183

The extent of MTB transmission among close and casual

contacts of two tuberculosis cases in two different high

schools was investigated. A considerable difference was

observed between the two cases in regard to the individuals

infected. In case A, 25 out of 152 close contacts and casual

contacts were infected and three were identified with secondary

tuberculosis. In contrast, no transmission was demonstratedin 121 casual and close contacts of case B, despite the fact

that only case B was smear positive. Given that no other

source of infection was identified, transmission of MTB in

case A was considered to have occurred within the school.

This is in accord with previous observations that suggest

contagiousness is not an all or nothing phenomenon and is

affected by several factors only one of which is the bacterio-

logic status of the patients sputum.

13

The environment and the clinical characteristics of each

case were considered as the most important factors in the

transmission variability.

1418

Several possible explanations

could be given to justify the difference in infectivity in these

two index cases. Case A occurred in winter when the

windows in schools are typically closed, resulting in poor

ventilation.

19

While case B occurred in the spring when all

classes were well ventilated. Furthermore, classmates in case

A were protesting against a new examination schedule and

they all gathered together in a classroom talking, singing, and

spending most of the day in close proximity. Moreover, there

are some clinical characteristics that could be used as possible

explanations for the observed difference. Case A complained

of fever and cough and was in contact with many students for

more than 20 days before he was administered preventive

chemotherapy. Case B also presented with cough and hemo-

ptysis but she started chemotherapy 4 days after the initiationof the symptoms, thus minimizing the period of transmission.

Host factors may also influence tuberculosiss epidemi-

ology. It seems that there is genetic variability (ethnic and

racial factors)

20,21

in human tuberculosis susceptibility (certain

Nramp1 alleles). In this case, we believe that ethnic and racial

factors should not be considered as possible risk factors as

the population was homogeneous.

Experimental evidence suggests that characteristics of the

pathogen may also contribute to the observed variability in

transmission. Remarkable variation in infectivity has been

observed in patients with similar clinical characteristics

22,23

although there are still conflicting data in the literature.

24,25

This possibility cannot be excluded, since the two isolates

belonged in two different strains, though we believe that the

different environmental and clinical factors contributed sig-

nificantly to the observed rates of transmission.

The evaluation of persons, who have been in contact with

index cases, is an important component of tuberculosis

control programs. When the index case concerns a student,

there is significant pressure to extend the screening to all

students of the school.

26

Moreover, the criteria used in our

study, to identify infected individuals among the BCG

vaccinated students, proved to be correct since no infected

students were identified within the control groups. Even

though, in countries where BCG vaccination status is used,

misinterpretation of results could occur. Thus, it is suggested

that screening is minimized to those contacts with increased

risk of infection. Due to our results (increased risk in closecontacts and less risk in casual contacts), case investigations

in schools should be done in two steps. Primarily, close

contacts should be examined and if any infected individuals

are identified, casual contacts should be included in the

screening process.

In conclusion, the case investigation of two pulmonary

tuberculosis cases in two high schools in an urban region of

central Greece revealed significant differences in the group

prevalence of infection between the two cases, which

presumably reflected environmental and clinical differences

related to the index cases.

References

1 Dolin PJ, Raviglione MC, Kochi A. Global tuberculosisincidence and mortality during 19992000. Bull. World HealthOrgan.

1994; 72

: 21320.2 Calpe JL, Chiner E, Sanchez E et al.

Micro-epidemics oftuberculosis; apropos of 2 school outbreaks in the area of theValencia community.Arch. Bronconeumol.

1997; 33

: 56671.3 World Health Organisation. Treatment of Tuberculosis.

Guidelines for National Programmes.

WHO, Geneva, 1993.(WHO/TB.91.161.)

4 Advisory Council for the Elimination of Tuberculosis ACET.The role of BCG vaccine in the prevention and control oftuberculosis in the United States. A Joint Statement by theAdvisory Council for the Elimination of Tuberculosis and theAdvisory Committee on Immunization Practices. MMWR

1996; 45

: 118.5 Ridzon R, Kent JH, Valway S et al.

Outbreak of drug-resistanttuberculosis with second-generation transmission in a highschool in California.J. Pediatr.

1997; 131

: 8638.6 Sacks JJ, Brenner ER, Breeden DC, Anders HM, Parker RL.

Epidemiology of tuberculosis outbreak in a South Carolinajunior high school.Am. J. Public Health

1985; 75

: 3615.7 Navarro Gracia JF, Pena Fernandez M, Garcia Abad I et al.

Tuberculosis outbreak at a public school.Rev. Clin. Esp.

1997;

197

: 1527.8 The Lodi Tuberculosis Working Group. A school and

community based outbreak ofMycobacterium tuberculosis

innorthern Italy, 19923.Epidemiol. Infect.

1994; 113

: 8393.9 Bredin CP, Godfrey M, McKiernan J. A school micro-

epidemic of tuberculosis. Thorax

1991; 46

: 9223.10 Querol JM, Oltra C, Minguez J et al.

Description of schoolmicro-epidemics of tuberculosis. Enferm. Infecc. Microbiol.Clin.

1993; 11

: 26770.11 Binkin NJ, Ghersi G, Boeri V, Lo Monaco R, Salamina G. An

epidemic of tuberculosis in an elementary school, Sanremo,Italy, 1993.Rev. Epidemiol. Sante Publique

1994; 42

: 13843.12 Rodriguez EM, Steinbart S, Shaulis G, Bur S, Dwyer DM.

Pulmonary tuberculosis in a high school student and a broad


5/6

184 C Hadjichristodoulou et al

.

contact investigation: lessons relearned.Md. Med. J.

1996; 45

:101922.

13 Menzies D. Issues in the management of patients with activepulmonary tuberculosis. Can. J. Public Health

1997; 88

: 197201.14 Hoge CW, Fisher L, Donnel HD Jr et al.

Risk factors for trans-mission ofMycobacterium tuberculosis

in a primary schooloutbreak: lack of racial difference in susceptibility to infection.

Am. J. Epidemiol.

1994; 139

: 52030.15 Houk VN, Baker JH, Sorenson K, Kent DC. The epidemiology

of tuberculosis infection in a closed environment. Arch.Environ. Health

1968; 16

: 2635.16 Rao VR, Joanes RF, Kilbane P, Galbraith NS. Outbreak of

tuberculosis after minimal exposure to infection. BMJ

1980;

281

: 1879.17 Braden CR. Infectiousness of a University student with

laryngeal and cavitary tuberculosis. Investigative Team. Clin.Infect. Dis.

1995; 21

: 56570.18 Frieden TR, Sherman LF, Maw KL et al.

A multi institutionaloutbreak of highly drug-resistant tuberculosis: epidemiologyand clinical outcomes.JAMA

1996; 276

: 122935.19 Toyota M, Morioka S. Tuberculosis outbreak in a junior high

school in Kochi City-studies on factors relating to extent of

tuberculosis infection and the efficacy of isoniazid chemo-prophylaxis. Kekkaku

2001; 76

: 62534.

20 Roberts CM, Musiska M. Results of an extended tuberculosisscreening programme among sixth formers in a Londonschool-more questions than answers. Commun. Dis. Public

Health

2003; 6

: 225.21 Kim SJ, Bai GH, Lee H et al.

Transmission of Mycobacteriumtuberculosis among high school students in Korea. Int. J.Tuberc. Lung Dis.

2001; 5

: 82430.22 Valway SE, Sanchez MP, Shinnick TF et al.

An outbreakinvolving extensive transmission of a virulent strain ofMyco-bacterium tuberculosis

.N. Engl. J. Med. 1998; 338: 6339.23 Bosley AR, George G, George M. Outbreak of pulmonary

tuberculosis in children.Lancet1986; 1: 11413.24 Bloom R, Small P. The evolving relation between humans and

Mycobacterium tuberculosis. N. Engl. J. Med. 1998; 338:6778.

25 Rhee JT, Piatek. AS, Small PM et al. Molecular epidemiologicevaluation of transmissibility and virulence of Mycobacteriumtuberculosis.J. Clin. Microbiol. 1999; 37: 176470.

26 Kobayashi H, Iriyama M, Amano T. Minor outbreak of Tuber-culosis infection in a junior high school-infection from apreventable case. Kekkaku 2003; 78: 61927.


6/6

mycobacterium tuberculosis transmission among high school

Documents