Abstract
Aims:
Pyrazinamide (PZA) is an important component of first-line tuberculosis (TB) treatment because of its distinctive capability to kill subpopulations of persister Mycobacterium tuberculosis (MTB). The significance of PZA can be understood by its inclusion in the most recent World Health Organization-recommended multidrug-resistant (MDR) TB regimen. Very little information is available about the prevalence of PZA-resistant TB from geographically distinct regions of high burden countries, including Khyber Pakhtunkhwa (KPK), Pakistan, because drug susceptibility testing (DST) of PZA is not regularly performed due to the complexity. In this study, we aimed to find the prevalence of PZA resistance in geographically distinct, Pashtun-dominant KPK Province of Pakistan and its correlation with other first- and second-line drug resistance.
Materials and Methods:
In this study, DST of PZA was performed through an automated BACTEC MGIT 960 system (BD Diagnostic Systems). The resistant samples were further subjected to DST of isoniazid (INH), rifampicin (RIF), ethambutol (EMB), streptomycin (SM), moxicillin (MOX), amikacin (AMK), ofloxacin (OFX), kanamycin (KM), and capreomycin (CAP).
Results:
Out of 1,075 MTB-positive isolates, 83 (7.7%) were found to be resistant to PZA. Among the PZA-resistant isolates, 76 (90–91.6%) and 67 (80–80.7%) were found to be resistant to INH and RIF, respectively, whereas 63 (76%) were resistant to both first-line drugs, INH and RIF (MDR-TB). The resistance level of EMB, OFX, and SM was also significantly high in PZA resistance, 35 (42%), 40 (48%), and 41 (49–50%) respectively.
Conclusion:
PZA resistance is significantly associated with other first- and second-line drug resistance. A significant number of PZA-resistant isolates are MDR cases. Therefore, DST of PZA should regularly be performed along with other drugs for better management of treatment of MDR and extensively drug resistant (XDR), to avoid side effects in patients.
Introduction
T
Mutations in pncA gene is the primary cause of PZA resistance. 8 However, a scattered type of mutation was reported in the entire pncA gene. This property makes it hard to develop a marker for rapid detection of PZA resistance. 9 In previous studies, the efflux rate of PZA/POA has also been suggested for predicting PZA resistance.10,11 In a study conducted by Doustdar et al., 10 PZA-resistant isolates lacked mutations in pncA, suggesting alternative mechanisms of PZA resistance.12,13 BACTEC MGIT 960 automated system for phenotypic PZA susceptibility testing in liquid culture medium is a well-developed method in practice. 14
Very little information is currently available on the prevalence of PZA resistance, because drug susceptibility testing (DST) of PZA is not regularly conducted in countries highly burdened by TB due to the complexity of PZA DST without WHO recommendation. The BACTEC MGIT 960 system PZA (BD Diagnostic Systems NJ) has now became the reference technique for DST because of its automated system. 14 Data are scarce on the prevalence of PZA resistance in Khyber Pakhtunkhwa (KPK) of Pakistan. This study aimed to find the prevalence of phenotypic PZA resistance and its association with other drug-resistant isolates in KPK Province, Pakistan.
Materials and Methods
Study design
This study was conducted between January 2016 and January 2017. All follow-up and diagnostic patients have been included and the data of patients have been collected from guardians or care takers.
Ethical considerations
This investigation has been evaluated and approved by the Institutional Ethics Committee of CUST Islamabad and Provincial Tuberculosis Reference Laboratory, KPK.
Study samples
A total of 8,119 samples have been referred by 17 Districts of TB Control Officers (DTO) and Programmatic Management of Drug-Resistant TB Units (PMDT) to the BSL-III Lab of Provincial Tuberculosis Reference Lab, Hayatabad Medical Complex (HMC), KPK, Pakistan.
Sample processing, isolation, and mycobacterial culture
Samples have been processed using N-acetyl-L-cysteine (NALC)–NaOH concentration method 15 by transferring to a falcon tube with equal volume of the NaOH/NALC, vortexed, and incubated at room temperature for 15 min for decontamination and digestion. A 50 ml phosphate buffer was transferred to each tube and centrifuged at 3,000 rpm for 15 min. The supernatant was then discarded in a container containing 5% phenol, while the pellet was mixed with phosphate buffer and cultured on LJ media and mycobacteria growth indicator tubes (MGIT) containing 7H9 media.
Culturing and identification of MTB
About 800 μl mycobacteria growth indicator tube, BD BACTEC™ MGIT™ (BBL-MGIT), a mycobacterial growth indicator tube barcoded for easy use with the automated BD BACTEC™ MGIT™ instrument for mycobacterial growth, detection and susceptibility testing, growth supplement and polymyxin B, amphotericin B, nalidixic acid, trimethoprim, and azlocillin (BBL MGIT PANTA) was added to the MGIT tube. A sample of 500 μl from the processed decontaminated specimen was also added to this tube. The MGIT tubes were kept in the MGIT 960 machine (BD Diagnostic Systems), which automatically senses growths in the tube within the recommended 42 days. The instrument was checked for a positive signal every day. In the case of a positive indication from the machine, the tubes were analyzed under a light, which on shaking showed small clumps, or cords like snowfall moving down toward the bottom of the tube. To confirm whether the growth is MTB, the BD MGIT MTBc identification test (TBc ID, Ref: 245159, Becton, Dickinson), a rapid chromatographic immunoassay, was performed, which detects the MTB complex antigen MPT64 secreted during culture of MTBc cells. Approximately 100 μl of the sample was taken from the MGIT-positive tube and added to TBc ID device well. Within 15 min, the MTB positivity was indicated by the emergence of pink to red at the Test “T” and the control “C” location, which has shown the recognition of antigen MPT64 of MTB in the sample. All the TBc-confirmed MTB tubes were subjected to susceptibility tests.
Drug susceptibility testing
DST of PZA was performed through an automated BACTEC MGIT 960 system (BD Diagnostic Systems). Samples were marked as PZA resistant if growth was found at 100 μg/ml of PZA critical concentration. The PZA-resistant samples were further subjected to phenotypic DST of INH, RIF, MOX, EMB, AMK, SM, CAP, OFX, and KM through BACTEC MGIT 960 system according to the policy guidelines of the WHO. 16 Growth of MTB at a critical concentration of drugs, 0.1 mg/ml INH, 1.0 mg/ml RIF was used to define resistance during sensitivity testing. 17 MTB H37Rv was used as a growth control (GC) along with DST.
Interpretation of DST
When the GC reached the growth unit (GU) value of 400 or more, the instrument indicated that the test was completed, and an inventory report was printed as “S” when the GU value was less than 100 or “R” when the GU value was greater than 100.
Data analysis
All the data of patients and drug resistance were entered using Epi-Data entry version 3.1 software and were analyzed through EPi-Data analysis software.
Results
Sociodemographic characteristics
A total of 1,075 MTB positive patients' samples were collected from 17 district tuberculosis officers (DTO) and PMDT of KPK. Of these, 587 (54.6%) were female and 488 (45.3%) were male (Table 1); 52 of these samples were extrapulmonary and 1,023 were pulmonary. Data regarding the history of TB in the participants were as follows: 410 (38%) were follow-up patients and 665 (62%) were diagnostic patients. The mean age of the patients was 36.5 (standard deviation ±18.8) (range: 4–100 years). None of these patients were HIV positive (Table 2).
TB, tuberculosis.
Drug susceptibility pattern
Using the BACTEC MGIT 960 system for drug susceptibility testing, a total of 83 (7.7%) isolates out of 1,075 MTB positive were classified as PZA resistant. The number of isolates resistant to PZA received from each TB center is illustrated in Table 3. Out of these 83 isolates, 34 (41%) were from male patients and 49 (59%) were from female patients. The resistant samples were further checked manually to confirm the growth of MTB against the critical concentration of drug. Out of 83 PZA-resistant samples, 76 (90–91.6%) and 67 (80–80.7%) were also INH and RIF resistant. Isolates that were resistant to both INH and RIF are called MDR, and were found in 63 (76%) of the PZA-resistant isolates. These results show that there is a strong association between PZA resistance and resistance to other drugs (Table 4).
DTO, Districts of TB Control Officers; PMDT, Programmatic Management of Drug-Resistant TB Units; PZA, pyrazinamide.
RIF, rifampin; INH, isoniazid; MOX, moxicillin; EMB, ethambutol; AMK, amikacin; SM, streptomycin; CAP, capreomycin; OFX, ofloxacin; KM, kanamycin.
Discussion
In our study, we found that the prevalence of PZA resistance in both diagnostic and follow-up patients in the KPK population was 7.7%, which seems to be high. These findings show that persistent bacilli, which are the main target of PZA, survived in resistance isolates, increasing the risk of drug resistance incidence of TB in these areas. Moreover, the resistance level of PZA increases with increase in resistance to other drugs, specially two important first-line drugs, INH (90–91.6%) and RIF (80–80.7%), whereas isolates that were found to be resistant to INH and also to RIF (MDR) were found in 63 (76%) of PZA-resistant samples. The resistance level of some other drugs in PZA-resistant isolates was as follows: ofloxacin 40 (48.2%) and streptomycin 41 (49.4%). We suggest that inclusion of PZA in a TB treatment regimen appears to be essential in susceptible cases; however, in the case of MDR, susceptibility of PZA must be performed prior to its inclusion in the treatment regimen as a large number of MDR cases are also resistant to PZA. In such conditions, it has less efficacy compared to the side effects of MDR-TB treatment. Therefore, PZA DST must be regularly performed along with other first-line drugs. Previous studies have shown a correlation between PZA and RIF resistance through molecular detection of mutations in pncA of MTB. 18 In a study by Zignol 2016, 4,972 patients were screened to find the prevalence of PZA resistance from five countries, the results of which varied from 3.0–42.1% in the surveyed settings, but were lowest in Pakistan, Bangladesh, and South Africa. However, the levels of resistance for all patients in all countries correlated significantly with rifampin resistance, except Pakistan where rifampicin resistance level is higher than PZA in all patients groups (p < 0.0001). 19
Previously, the resistance level of PZA in MDR patients has been calculated as 37 of 71 (52.1%), while 6 of 59 (10.2%) among fully susceptible cases. 20 These findings suggest that there is approximately half a chance to successfully eradicate persistent bacilli in MDR patients. Therefore, it is suggested that PZA DST should be regularly performed and PZA inclusion to treat MDR should be linked after susceptibility test results are made available. More studies with a large sample size should be conducted for better estimation of PZA prevalence in the KPK population of Pakistan.
The correlation of PZA resistance to other drugs should be investigated genotypically to confirm molecular correlation by mutations. Genotypic methods of resistance correlation in MDR will further support these observations for better management of TB treatment. 21
Conclusion
The prevalence of PZA resistance is significantly associated with resistance to isoniazid, rifampicin, streptomycin, and ofloxacin. A large number of PZA-resistant isolates are also MDR and extensively drug resistant (XDR) cases, which affect the treatment outcomes of persister MTB and increase the chances of latent TB. Therefore, PZA inclusion to treat MDR should only be done after conducting susceptibility testing. Consideration should also be given when defining the role of PZA in future TB treatment regimens. We suggest that PZA susceptibility testing should regularly be performed along with other frequently used drugs for better management of TB treatment.
Footnotes
Disclosure Statement
No competing financial interests exist.