- To define the risk factors for TB mortality in children, comparing children in the houses with and without adult TB cases, controlling for determinants of transmission and for other determinants of mortality
- To investigate the impact of chemoprophylaxis on TB morbidity and mortality among children exposed to adult TB cases.
- To assess if unsupervised isoniazid (IHN) preventive therapy (IPT) is feasible in a setting like Guinea-Bissau
Tuberculosis remains a major public health problem in Guinea-Bissau. The current epidemic of HIV-2 infection and the emerging epidemic of HIV-1 and dual infections (HIV2 + HIV1) are likely to rapidly increase the spread of TB in the community.
Epidemiological features of childhood tuberculosis in Guinea-Bissau are unknown.
From 1993 to 1998, 250 cases of childhood TB were diagnosed at the referral centre of the TB Programme, Hospital “Raoul Follereau”, in Bissau (Annual reports). Among these, only 13% were smear positive for acid fast bacilli (AFB). For the country, the number of deaths associated with TB in children in Guinea-Bissau is not known. However, between 1993 and 1998 the “Raoul Follereau” Hospital registered 21 (8.4%) deaths from TB out of 250 hospitalised children less than 15 years of age (Annual reports).
During the community study of TB, which did not focus on children, we found a low incidence of TB among children. Identifying children with TB is difficult since the majority of them are diagnosed clinically rather than by isolation of M. tuberculosis. Children are most likely to be infected with TB as a result of exposure to an adult with pulmonary TB, hence the rate of infection and disease will be highest among children in close contact with an index case. Identification of TB in children will thus require a case contact investigation in households.
The study was conducted from September 2005 to July 2008 in six areas covered by the Bandim Health Project (BHP), a demographic surveillance site part of the International Network for the demographic evaluation of populations and their health (INDEPTH) in developing countries network. The current population of 102,000 is followed through regular censuses and registered with information on sex, ethnic background, date of birth, death and migration. Information regarding hospitalisations and deaths are collected every 3 months for children under 3 years of age. All paediatric hospitalisations from the study area have been registered since 1990
Since 1996 a TB surveillance system, implemented in collaboration with the national TB hospital (Hospital “Hospital Raoul Follereau”) has identified adult TB cases using passive and active case finding. Adults permanently or temporarily living in the study area presenting at any of the three health centres with symptoms or signs of active TB were referred to the national hospital for further investigation and possible inclusion. Furthermore, nurses visited houses of TB cases in order to detect secondary cases and referred suspect cases to the TB hospital. From the above mentioned period around 170 TB cases have been registered each year, 2/3 being residents in the area and 1/3 visitors coming from the rural areas to be treated in Bissau.
Participants and recruitment
Between September 2005 and October 2007 we enrolled of the children less than 15 years old on IPT study in the study area. Once an adult from the study area was identified with pulmonary TB a project assistant went to the patient’s house and updated the census. Houses are multifamily dwellings with an average of 3 households per house. Children less than 15 years old living in the house when the adult started treatment were eligible for inclusion; children less than 5 years old were eligible for IPT regardless of TST status while children 5-15 years of age were eligible if the mean of the longitudinal and transversal indurations of the TST reaction was >10 millimetres. Children were enrolled in the study and put on IPT if a parent or person in loco parentis gave consent. Prior to the initiation of INH prophylaxis the children were evaluated for active TB in a clinical examination for signs and symptoms of TB. For children under 5 years of age a “weight by age chart” and the Keith Edwards score were applied.
Preventive therapy, INH administration, adherence and completion
INH tablets were administered at 5 mg/kg daily together with Pyridoxine (Vitamin B6) tablets. The Vitamin B6 dosage was 25 mg for children receiving <100 mg of INH and 50 mg for children receiving >100 mg of INH. The medicine was provided every two weeks by a field assistant. Pill counts were performed at each visit for distribution of meidicines and evaluated on a monthly basis. Only the biweekly number of missed doses was recorded; the exact dates when a child failed to take a dose was not obtained. To assess the continuity of the drug consumption we looked at the number of completed months of treatment.
Follow-up and side-effects
The children were visited 4 times during treatment period. The evaluation at follow-up visits included questions about side effects and physical assessment of signs and symptoms of hepatotoxicity, e.g. jaundice and vomiting. Weight was registered and entered on a weight-by-age chart in order to detect failure to thrive or malnutrition during IPT. Routine laboratory monitoring of transaminase levels during treatment was not performed unless a child presented signs or symptoms suggestive of hepatotoxicity.
Pre-IPT cohort (Mortality among exposed children)
In this study mortality in children less than 5 years of age living with an adult TB from the period 1996 to 1998 was compared to the mortality of children in the general population.
TB diagnose and inclusion procedures
Criteria for inclusion of TB cases in the study was age >15 years and one or more of the following symptoms and signs without other explanatory disease; cough more than one month without improvement after antibiotic treatment, fever constantly or periodical for more than one month, weight loss, dyspnoea, haemoptysis, night sweats or lymphadenopathy. Patients were examined clinically, interviewed using standardized questionnaires, direct microscopy of morning sputum was done on 3 consecutive days as well as sputum culture and tuberculin skin test (using Multitest®, Bio-Merieux, France). Frontal chest x-rays were performed and pathological findings in the chest, including pulmonary changes, pleuritis and hilar lymph node enlargements, were regarded as intrathoracic manifestations. Blood was drawn for HIV-test. Suspected cases with one or more sputum smears positive at direct microscopy or positive in culture for M. tuberculosis were regarded as tuberculosis. Suspected cases with one or more sputum smears positive for acid-fast bacilli in direct microscopy were regarded as smear-positive tuberculosis. Patients with clinical signs, symptoms and x-ray changes compatible with active intrathoracic tuberculosis, but without positive bacteriological test, were treated with antibiotics (co-trimoxazole or amoxicillin) and then re-evaluated clinically and with chest x-ray. If there was no improvement and suspicion remained, the patient was diagnosed as having presumed tuberculosis. By definition, intrathoracic TB includes both pulmonary and extrapulmonary TB (in our case pleural effusion and intrathoracic lymphadenopathy). Patient diagnosed as having intrathoracic TB but unable to produce sputum samples have been coded as smear-negative.
IPT cohort (impact of IPT on mortality)
The IPT cohort included children enrolled on IPT between September 2005 and October 2007. Once an adult from the health centres within the study area was identified with pulmonary TB a project assistant went to the patient’s house and updated the census for the families living in the index-house, and socio-economic and demographic information was noted on questionnaires.
Children less than 5 years of age living in the house, when the adult with TB started treatment, were eligible for inclusion. Children were enrolled in the study and put on IPT if a parent or caregiver gave consent. Prior to the initiation of IPT, active TB was excluded in a clinical examination for signs and symptoms of TB. For children less than 5 years of age a “weight by age chart” and the Keith Edwards score were applied. In children who presented fever and respiratory symptoms inclusion was delayed until active TB was ruled out. For children in whom malaria was diagnosed anti-malarial treatment was administered according to the local guidelines. Children who presented signs and symptoms suggestive of TB while on IPT were submitted careful and thorough assessment of all evidence from a careful history, clinical examination and relevant investigations, e.g. laboratory examination including HIV testing and chest x-ray. Broad spectrum antibiotics were administered for 10-15 days. Children who failed to improve clinically and radiologicaly after 2 weeks of broad spectrum antibiotics, and without other explanatory disease were given a full TB treatment regimen according to the national protocol. The cohort and study routines are described in details elsewhere (Paper1)
A total of 2373 children were identified as contacts of 304 adult TB cases. Among the identified children, 1943 (82%) were evaluated for eligibility to IPT with a total of 820 children being enrolled in the study, 603 children <5 years and 217 children 5-15 years old were enrolled on IPT. In total, 79% of the prescribed doses were taken with 65% of the children taking more than 80% of the prescribed doses. In all 77% of the children completed more than 6 months of treatment with at least 80% adherence, 50% completed more than 6 consecutive months of IPT. Moving out of the study area and traveling accounted for 59% of the missed doses. Adverse events to IPT were few and mild.
Mortality among exposed children
In this study we retrospectively compared mortality in children less than 5 years old exposed to tuberculosis at home with the mortality among children with unknown exposure in the general population. The study provided several important observations. First, we observed 66% higher mortality among the exposed children. The lack of excess mortality in the selected houses 3 years prior to exposure indicates that the increased mortality is likely to be associated with TB. Second, there was a strong association between exposure time and mortality, with a startling 7-fold higher mortality 12 months after exposure for the oldest children. Third, excess mortality was 2-fold higher mortality among children living in the same household as the adult TB case compared to the children in the houses without exposure at home and it was 8-fold increased for children of mothers with TB. Fourth, mortality was highest in children exposed to an adult with smear positive pulmonary TB but those exposed to a smear negative TB case had also significantly increased mortality.
Impact of IPT on mortality
In the present study we have shown a considerable impact of IPT on mortality among children less than 5 years of age exposed to adult TB. Children exposed to TB in 1996-1998, when IPT was not available in the area, suffered a 66% excess mortality compared to unexposed children. This excess mortality was completely removed in the 2005-2007 cohort of children receiving IPT. Unexpected observations were that mortality in the study area declined dramatically between the two study periods and mortality declined more than expected among both exposed children receiving IPT and exposed children not receiving IPT. This may suggest that some of the excess mortality associated with TB exposure is caused by interactions with other childhood infections and if these go down TB-associated mortality may also decline.
Overall adherence to IPT was generally high in Guinea Bissau but only half of the included children completed at least 6 consecutive months of treatment with more than 80% adherence. Migration and traveling accounted for most of the observed non-adherence.
The present study on impact of exposure to TB shows that intimate house contact with an adult TB case increases the risk of death considerably among children less than five years of age. Although the excess mortality was higher when the adult TB case had sputum smear positive pulmonary tuberculosis there was also a significantly increased risk among children exposed to smear negative adult TB cases. In our analysis there were around 16 excess child deaths among TB exposed children. It is important to note that excess mortality only started from around 6 months after exposure. Hence, there should be ample time for implementing preventive measures.
Our study clearly shows that children less than 5 years of age exposed to TB at home have a high mortality that can be prevented with IPT. All our data indicate that IPT should be part of the standard TB programme, and would have a large impact on child mortality in low income countries.