Abstract:
The resurgence of tuberculosis (TB) has been associated mainly with Human Immuno
deficiency Virus (HIV) and widespread poverty. The HIV pandemic and emergence of
resistant strains of Mycobacterium tuberculosis threatens the success of patient treatment
and TB control programs. The initial treatment of TB involves the administration of four
drugs namely; isoniazid, rifampicin, pyrazinamide and either ethambutol or
streptomycin. Isoniazid and rifampicin are critical in successfull treatment of TB.
Resistance to any anti- TB drugs can develop spontaneously as a result of mutation then
there after under selective pressure. Few epidemiological studies done in Nairobi have
shown the presence of multidrug resistant TB (MDR TB) but none has determined the
distribution of these cases and their potential causes. There is also limited information
on the exact prevalence of resistance to anti-tuberculosis drugs in high density
populations with high rates of tuberculosis and HIV co-infection such as those in Nairobi
where there is great potential for spread. The objective of this study was to determine the
magnitude, strain patterns and and identification risk factors associated with the
resistance. This was a cross sectional study. Study sites were selected from each of the 8
districts in Nairobi and study subjects selected using weighted cluster sampling based
on the patient population proportions of new smear positive cases notified in 2007~
Eligible patients were enrolled in the study consecutively after consenting until the
number allocated to each study site was achieved. Sputum samples were collected from
patients before start of treatment in accordance with standard programme procedures
XVI
Image
and transported to the central reference laboratory (CRL) where they were processed
for culture and drug susceptability testing (DST) using the standard procedures. All
culture isolates positive for Mycobacterium tuberculosis (MTB) were tested for
susceptibility to first line anti-tuberculosis drugs namely; isoniazid, streptomycin;
rifampicin and ethambutol using the conventional Mycobacterium Growth Indicator
Tube (MGIT) medium. Spoligotying molecular method was used to identify isolates
with strain families. Data was analysed using the SPSS software version 17. The
standard chi-square test and Fishers exact test were used to compare 2 data points and a
p-value of less than 0.05 was considered significant. The Mantel-Hansen test was used
for odds ratios. A total of 691 TB patients were enrolled for study with mean age of 31.3
and median of 30 years. Among the patients, 438 (63.4%) were new smear positive TB
cases from whom 368 Mycobacterium tuberculosis isolates were subjected to drug
susceptiblity testing (DST) and 253 (36.6%) were previously treated smear positive TB
cases from whom 199 MTB isolates were subjected to DST. Eeighty five percent and
seventy seven percent of the MTB isolates from new patients and previously treated
patients, respectively, were fully sensitive to all the drugs tested. Initial resistance (i.e.
resistance among new patients) to isoniazid, streptomycin, ethambutol and rifampicin
was 10.3%, 4.3%, 5.1 % and 0.81 %, respectively. Among previously treated patients
resistance to isoniazid, streptomycin, ethambutol and rifampicin was 18.1%, 10.5%,
7.03% and 9.04%, respectively. The prevalence of MDR TB was 0.54% and 8.54%
among new and previously treated patients respectively. No Extensive drug resistant TB
(XDR TB) was identified in this population. The study found out that the TB disease in
XVll
Nairobi was caused by five main MTB strain sub-families namely; CASl_KIU, Tl,
Beijing, LAM9 and LAM3 and S/Convergent which collectively contributed to 57 % of
the TB cases in Nairobi. The levels of drug resistant TB in Nairobi was high compared to
other previous studies done in the country and there was diverse array of Mycobacterium
tuberculosis strain families which could be indicative of a cosmopolitant population with
frequent migration suggesting that the dorminant strain families may have been present in
the population for an extended period of time or on going transmision of closely related
strain families. HIV status of the patient was not associated with any drug resistance. The
proportion of isolates from patients who had previous treatment history (24.3%) was
significantly higher than those from new patients (15.1%). History of previous treatment
was therefore strongly associated with drug resistance (p=O.007 for any resistance and
p<O.OOOl for MDR TB). The study did not find any association between drug resistance
and the Beijing or any other MTB strain family. However the overall proportional
contribution to the TB disease in Nairobi by the Beijing strain family alone had increased
from 8% to 12% over the last 5 years. This poses a serious threat to TB control due to its
high virulence and frequent association with multidrug resistance. The National TB
control program should therefore rapidly intensify the implementation of services for the
diagnosis and treatment of MDR TB to control its transmission and carefully monitor the
trends of anti-tuberculosis drug resistance to ensure the success of the TB control
programme.