Epidemiology of tuberculosis in humans in Narok District with special interest on mycobacterium bovis
Abstract
Bovine Tuberculosis (BTB) is an important zoonotic disease among most
pastoralist communities in sub-Saharan Africa. A possible reason for this is the close
association pastoralists have with livestock and their habit of consuming raw milk.
The main objective of this study was to estimate the proportional morbidity of
zoonotic BTB caused by M bovis from clinical human tuberculosis (TB) cases and to
identify the risk factors associated with both M bovis and M tuberculosis infections
in Narok District, Kenya, with a focus on the role of raw milk consumption as a risk
factor.
The study first describes morbidity patterns from past medical records on
human pulmonary tuberculosis (PTB) and extra-pulmonary tuberculosis (EXPTB)
cases across the country (EXPTB cases are usually more likely to be associated with
BTB than other tuberculoses). 'Secondly, past morbidity data from hospital records in
Narok District covering both forms of TB were analysed in more detail. Thirdly, a
survey questionnaire was administered to suspect TB patients reporting to 17 health
units in Narok District. The questionnaire was designed to gather information on risk
factors for cattle-to-human transmission of BTB. In addition, sputum and/or cervical
lymph node aspirate samples from the suspect patients were obtained for laboratory
identification by acid-fast staining (AFS), culture and biochemical tests to isolate the
Mycobacteriacae.
Available hospital records showed that approximately 180,000 TB patients
were diagnosed by AFS nationally between 1990 and 1999. The national overall ratio
of human PTB to EXPTB for the decade was 4: 1. There was a general increase in
reported cases of both forms of TB over the period (with a more marked increase
between 1996 and 1999), the general rate of which cannot be explained by population
growth or increase in number of hospital visits alone: EXPTB cases increased by 26%
annually, significantly higher than the rate of increase of PTB cases (17%). EXPTB
occurrence nationally was uncorrelated with production system or population
densities of either cattle or humans.
There were about 120,000 outpatient-visits in Narok District in 1999, similar
to the figures of the previous nine years. In the same year, clinical diagnoses of
human TB in the district were 2,603. Of these, 225 (9%) were confirmed acid-fast
positive (AFB+), indicating the presence of tubercle bacilli. Most (83%) of the AFB+
patients had pulmonary symptoms. The rest (17%) of the patients presented extrapulmonary
symptoms with or without pulmonary symptoms.
A total of 159 patients (135 from the lowland extensive pastoral area and 24
from the highland sedentary area) were clinically diagnosed and sampled for TB
during the nine-month period of study from April to December 2000 in Narok
District. Of these, 132 (83%) were positive on acid-fast bacilli testing. On culture,
only 32 (24%) out of the l32 samples yielded Mycobacterium tuberculosis.
No Mycobacterium bovis was isolated. The sampling accuracy implies a 95%
confidence that TB cases in the district that are bovine are not more than 2%. The rest
100 AFB+ cases (76% of the 132 tubercle bacilli positive samples) were not speciated.
These likely represented atypical and non-pathogenic tubercle bacilli that were neither
M tuberculosis nor M bovis.
Most patients (68%) sought health services in private/mission health units
compared to Government health facilities. Coughing was the best predictor of TB
infection with 75% sensitivity, 67% specificity and positive predictive value of
(PV+ve) of 92% for AFB+ result. Significantly more samples from patients living in
the lowland pastoral areas were AFB+ (91%) compared to samples from patients
living in the highland areas (38%). Descriptive results showed that most households
(hhs) that kept cattle also consumed raw blood and raw beef(p < 0.01). Consumption
of raw milk by 33% of patients was protective for human TB (p < 0.01), but
consumption of other raw animal products (blood or beef) was not (p > 0.1). Marital
status, age, sex, previous exposure to same infection as clinically perceived by the
patient and family size were not associated with TB infection (p > 0.1).
The study concludes that though these findings vindicate the long held official
position indicating absence of bovine tuberculosis in Kenya, the findings need to be
verified in other areas and periodically, given the risk posed by frequent movement of
pastoral livestock across borders from neighbouring countries. If M bovis were
present in Kenya, those at the greatest risk of acquiring BTB would be those people
who consume raw milk and/or other raw animal products. It appears from this study
that raw milk consumption -and possible ingestion of atypical non-pathogenic
Mycobacteriaceae that may be present in raw milk offers some protective immunity
against M tuberculosis infection. However, this benefit is likely to be outweighed by
the risk of infection from other pathogens such as brucellae that can be potentially
transmitted through consumption of raw milk. Boiling of raw milk is therefore
recommended before it is consumed.
Citation
Degree of Master of Science in Veterinary Epidemiology and EconomicsPublisher
University of Nairobi Department of Public Health, Pharmacology & Toxicology
Description
A thesis submitted in partial fulfilment of the requirement for the award
of the degree of Master of Science in Veterinary Epidemiology and
Economics,
Department of Public Health, Pharmacology & Toxicology,
Faculty of Veterinary Medicine,
University of Nairobi.