Trends of Xpert MTB/RIF in the diagnosis of Mycobacterium tuberculosis and rifampicin resistance in Southwest Nigeria: A 4-year retrospective study

INTRODUCTION

Tuberculosis (TB) is a communicable disease caused by the bacterium, Mycobacterium tuberculosis (MTB).^[1] It remains a global public health challenge, posing a growing concern in the developing countries, Nigeria inclusive.^[2-5] The increasing burden of drug-resistant TB (DR-TB), which can manifest as isoniazid-resistant TB, rifampicin-resistant TB (RR-TB), multidrug-resistant TB (MDR-TB), pre-extensively DR-TB, and extensively DR-TB, is particularly concerning.^[3]

In 2019, the World Health Organization (WHO) global TB report indicated that 1.7 billion people are at risk of TB, with an estimated 10 million new cases every year.^[3] Although notification of TB cases appears to be improving from 6.4 million new cases of TB in 2017–7.0 million in 2018, the actual burden of TB in the developing countries remains largely unknown.^[3-5] The suggested reasons include under-reporting and challenges with diagnosis, as two in every three cases of DR-TB go undiagnosed and thus are not reported.^[3]

In most developing countries, TB is still diagnosed based on the demonstration of the presence of acid fast bacilli (AFB) in clinical samples, such as sputum stained with Ziehl– Neelsen (ZN) staining.^[6] The microscopic confirmation of MTB in such smears is observer dependent and useful mainly in advanced disease. Furthermore, high false-negative results are typical with a small quantity of the organism and sensitivity range from 6% to 80%.^[7] In contrast, rapid molecular tests using the Xpert MTB/RIF have a sensitivity of 98.6% for MTB, reduced overall turnaround time to as low as 2 h compared with AFB smear which may last 3 days and Lowenstein-Jensen culture, 6 weeks.^[7,8] The Xpert MTB/RIF test demonstrated a high diagnostic yield for TB.^[9] Among the rapid molecular tests recommended by WHO, the Xpert MTB/RIF, which runs on Cepheid’s Xpert automated polymerase chain reaction (PCR) platform is preferred. The rapid turnover of 2 h, lack of need for extensive training for personnel conducting the test, and higher sensitivity and specificity in detection of MTB and rifampicin resistance TB make the Xpert MTB/RIF^® the preferred choice for diagnosis of TB based on the WHO recommendations.^[7,10]

Globally, the trend of MTB diagnosis has increased with a surge between 2017 and 2019.^[11] A similar pattern of sharply rising disease notifications was observed in Africa, Southeast Asia, and Latin America, albeit to a lesser extent.^[12] This improved notification has been attributed to increased utilization of molecular rapid molecular tests and other newer testing modalities.^[2,4]

In addition to MTB notification trends, monitoring anti-TB drug resistance is essential for achieving the 2035 END TB goal. This is accomplished through routine diagnostic drug susceptibility testing utilizing rapid molecular tests, culture methods, or sequencing technologies.

According to a recent systematic review and meta-analysis conducted by Onyedum et al.^[13] in 2017, the prevalence of DR-TB among newly diagnosed cases in Nigeria was 32%, while the prevalence rate among previously treated cases was 53.0%. The reported prevalence of 32% is significantly higher than the global prevalence of 3.4% reported for new cases in the 2019 WHO global TB report, reaffirming Nigeria as one of the high burden countries for DR-TB.^[2,9] Undernutrition, diabetes, smoking, harmful alcohol consumption, poor treatment adherence to anti-TB drugs; previous haphazard or poor implementation of isoniazid prophylaxis and previous ineffective TB treatment, especially in private practice; increasing treatment failure and loss to follow-up, poor Xpert MTB/RIF diagnostic coverage; and a high burden of human immunodeficiency virus (HIV) coinfection in low-resource settings, including Nigeria, are potential drivers for the high prevalence of resistant TB.^[13-22] Higher prevalence of social determinants such as poverty, inequality, unsafe housing, discrimination, and social stigma in low-resource settings such as Nigeria also plays a significant role in the spread of DR-TB.^[13,14,20]

According to the 2019 WHO global TB report, Nigeria has one of the highest burdens of MDR/RR-TB, accounting for 4.2% of new cases and 15% of previously treated MTB.^[2] The reported figure is comparable to those from South Africa, but higher than the regional average and lower than Somalia’s notification rate of 8.7% from new cases. Southeast Asia has roughly half as many MDR/RR-TB cases from new and old cases as Nigeria.^[4] While a higher proportion of MDR/RRTB in newly diagnosed cases was observed in Nigeria, disease notification of MDR/RR-TB in previously treated TB cases in the Americas and Nigeria was identical.^[2]

In terms of global trends, there appears to be a reassuring reduction in MDR/RR-TB cases between 2015 and 2018. The 2019 WHO global report indicates a 10% decrease in MDR/RR-TB cases with a wide confidence interval (CI). While countries such as Myanmar, Peru, and the Russian Federation experienced an increase, Zambia in Africa and Kyrgyzstan in Asia have observed a decline of approximately 3% per year.

As the world strives to meet the high-level UN TB treatment targets and the WHO End TB Strategy by 2035 benchmarks, the use of rapid molecular tests for early diagnosis and sensitivity testing is critical to achieving these goals, in addition to addressing the disease’s social determinants.

This study, therefore, aimed at tracking the trends of MTB and the yield of rifampicin resistance TB using Xpert MTB/RIF over a 4-year period at a tertiary health facility in Southwest Nigeria. It is hoped that the results would contribute to MTB surveillance and serve as an indicator of the TB program’s performance.

MATERIALS AND METHODS

RESULTS

Of the 4625 presumed cases evaluated for TB using the Xpert MTB/RIF test, we excluded 99 (2.1%) because of incomplete data. The flow diagram for participant enrolment is detailed in [Figure 1]. The mean age (standard deviations) of the presumed TB cases was 43.4 (18.8) years, with a range from 3 months to 102 years. Most of the presumptive TB cases were in the age group, 31–45 (30.6%) while children and adolescents (<18 years) were 349 (7.7%). Males were 2247 (49.6%), as shown in [Table 2]. Of the 4625 presumptive TB cases, 756 (16.7%) were living with HIV infection, 2443 (54.0%) were HIV uninfected, and in the remaining 1327 (29.3%), their HIV status was unknown. Based on the distribution, most presumptive TB cases 4436 (98.0%) were new TB cases (no prior TB diagnosis nor treatment), 50 (1.1%) were cases of TB relapse, 16 (0.4%) were TB treatment failure cases, and 24 (0.5%) had a history of contact with DR-TB cases [Table 2].

Close

Figure 1:

Flowchart for screening of patients with presumed tuberculosis and rifampicin resistance.

Export to PPT

The overall yield of MTB by Xpert MTB/RIF assay was 12.9% (584/4526). The yield was comparable in males (13.5%; 302/2247) and females (12.4%; 282/2279), χ² = 1.145, P = 0.288). Based on the age group, the highest yield of MTB was detected in the age 18–30 years (16.2%), while the least was in age <18 years (6.9%; P = 0.002), as shown in [Table 3]. Age <18 years was less likely than adults to have an MTB detected by Xpert MTB/RIF AOR = 0.416, (95% CI: 0.236, 0.732). The diagnosis of MTB using Xpert MTB/RIF was associated with HIV status, P < 0.001 [Table 3]. Those living with HIV infection had a lower yield of MTB by Xpert MTB/RIF, odds ratio (OR) = 0.398 (95% CI: 0.266, 0.700); AOR = 0.556, (95% CI: 0.408, 0.758) compared to being HIV uninfected or having an unknown status. The MTB detection rate was highest in those with TB treatment failure (18.8%; 3/16). There were no significant differences in the MTB detection rate based on the TB status, P = 0.407 [Table 3].

The detection rate of MTB increased from 5.6% in 2016 to 23.8% in 2019, P < 0.001 [Figure 2]. The overall proportion of RR-TB cases was 7.5% (95% CI: 5.50, 9.94). [Figure 3] shows that RR-TB cases detected declined from 28% in 2016 to 4.6% in 2019, P < 0.001. Bivariate analysis [Table 4] showed that age >45 years was significantly associated with RR-TB detection status with an OR = 1.968, (95% CI: 1.011, 3.830) and an AOR = 2.046, (95% CI: 1.046, 4.004).

Close

Figure 2:

Trend of prevalence Mycobacterium tuberculosis.

Export to PPT

Close

Figure 3:

Trend of rifampicin resistance among the confirmed Mycobacterium tuberculosis.

Export to PPT

Forty-two of the 44 cases of the RR-TB (42/572; 7.3%, [95% CI: 5.31, 9.75]) were new cases, with TB treatment failure and relapse cases accounting for 10.1% (2/11; 10.1%, [95% CI: 0.35, 42.5]) of all RR-TB cases (P < 0.01).

[Figure 4] captured the geographical distribution pattern of MTB and RR-TB detection. RR-TB appears to be more common in densely populated semi-urban areas than in rural areas. The RR-TB appears to be more prevalent in communities where non-indigenous migrants from other states in Nigeria with higher HIV infection rates have settled.^[27]

Close

Figure 4:

Map of the density and distribution of Mycobacterium tuberculosis detected by GeneXpert and the rifampicin-resistant tuberculosis.

Export to PPT

DISCUSSION

Given the high prevalence of MTB and RR-TB in Nigeria, this study examined the trends in MTB and RR-TB detection at TB treatment centers in Ogbomosho Southwest Nigeria using the Xpert MTB/RIF assay. In a large cohort of 4625 patients with presumed TB, the TB detection rate using a rapid nucleic acid-based assay (Xpert MTB/RIF) was 12.9%. The proportion of MTB cases detected is lower compared to 18.9% in Nasarawa State,^[28] North-Central Nigeria, 37.7% in Lagos,^[29] Southwest Nigeria, 17% in Benue,^[30] North-Central Nigeria, and 16% in Kebbi^[31] Northwest Nigeria. Although the studies mentioned above were all hospital based at referral TB treatment centers, detection of MTB using the Xpert MTB/ RR was less likely in individuals living with HIV infection which is in contrast to the higher yield among TB-HIV coinfected individuals in Benue and Nasarawa states. This might be related to the higher prevalence of HIV infection in these areas and concomitantly increased TB prevalence in the population with improved detection in the context of the use of Xpert MTB/RIF as the primary diagnostic tool.

Notably, the observed TB detection in the present study was higher than 9% obtained a decade earlier in Ibadan,^[32] Southwest Nigeria. The present report is also higher than the report from Port Harcourt (4.5%)^[33] in South-South Nigeria. These may be due to the use of the Xpert MTB/RIF for TB detection in the present study, as opposed to acid-fast bacilli (AFB) with ZN staining in the previous studies in Ibadan and Port Harcourt, South-South Nigeria. The Xpert MTB/RIF is known to have a higher sensitivity than traditional AFB with ZN stain. Furthermore, there are differences in the timelines and study durations between the current review and the preceding studies, with the majority of earlier studies lasting only 1 year.^{[28,29,31,33]} In addition, this may reflect regional differences in the burden of TB in the country, with densely populated areas such as Lagos state having a higher burden than semi-urban locations.

Studies from other countries, on the other hand, reported a significantly higher detection rate of MTB, with centers in Nepal, Ethiopia, and Iraq reporting 23.6%, 30.5%, and 40%, respectively.^[34-36] The preponderance of new presumed TB cases from the general population in the present study may account for the lower proportion of MTB cases compared to the report from Nepal, in which the denominator was comprised of high risk, secondary level screening of treatment failure cases, HIV-infected patients, and patients with radiographic evidence of TB who were all smear negative for the ZN acid-fast stain technique.^[37]

This study observed that the proportion of patients tested with Xpert MTB/RIF increased over the 4-year study period, as did the yield of MTB cases detected. This finding corroborates the global trend in case notifications of people diagnosed with TB, from 2016 with peak notification between 2017 and 2019.^[2-5] The detection of MTB in Southeast Asia followed a similar pattern to that observed in Africa. However, the magnitude of increased detection and notification was significantly less pronounced in the Eastern Mediterranean region, and a decline was observed in the Western Pacific region of the world in 2017.^[11] The increased MTB notification in African region may be a consequence of the increased access to Xpert MTB/RIF, a more sensitive diagnostic tool than the ZN acid-fast stain technique.^[7] In addition, throughout the years, screening criteria have been modified to allow for the testing of more people for MTB utilizing the Xpert MTB/RIF approach.

Notification of DR-TB is one of the surrogate measures of the effectiveness of global TB control programs, with varying reports based on diagnostic and treatment coverage. The present study’s finding of 7.5% (95% CI: 5.50, 9.94) RR-TB notification compares favorably with the published national average but is higher than the global average reported by the WHO in the 2019 report.^[2] Our finding is lower than the WHO reported incidence in India, North Thailand, China, and the Russian Federation. A rifampicin resistance rate of more than 3% in non-MDR-TB is a poor predictor of TB mitigation strategies.^[2,3,38-40] This finding has implications for the unfolding of the COVID-19 pandemic, where lockdowns and restricted access to care could be a driving force for the rapid spread of RR-TB, especially in settings with limited resources.^[11] Thus, the need for more pragmatic approaches, including ongoing surveillance for resistance cannot be overemphasize as the world strives to achieve the third Sustainable Development Goal 3 target of eradicating TB by 2030.^[2]

The observed 7.3% (95% CI: 5.31, 9.75) RR-TB among new cases compares to the national average of 4.3% (95% CI: 3.2– 5.5) in 2019 and the reported average of 8.6% in Eswatini, which is located in the East of South Africa. However, the percentage is lower than the global and African regional averages, as well as the Southeast Asian and Western Pacific averages of 3.4%, 2.5%, 2.6%, and 4.6%, respectively. In the present study, the reported percentage of 10.1 (95% CI: 0.35, 42.5) RR-TB among previously treated patients is in tandem across all of the above-mentioned regions and the global average.^[2]

Age over 45 is a significant risk factor for developing RR-TB as well as a poor predictor of treatment outcome.^[41] According to research from Africa, China, and Thailand, being over 45 years old is associated with a lower treatment success rate.^[42-44] Older age was associated with higher comorbidity with increased pill burden and adverse events and poor treatment adherence. This age group is also predisposed to interact with MDR-TB patients, and has a higher risk of smoking, alcoholism, loss of follow-up, and other social factors that contribute to the development of resistant TB. In this subpopulation, a targeted approach different from DOTS may be more relevant.^[45]

A reducing trend of RR-TB detection among confirmed MTB cases was observed over the study period. This is consistent with the trends of the global burden of MDR-TB between 1990 and 2017.^[46] The declining RR-TB cases in this study could be credited to Nigeria’s ongoing effective National TB and HIV control programs, as well as the availability of effective anti-TB drugs. Our center also has an effective contact tracing protocol and effective adherence counseling, as well as a holistic approach to a tripartite nature of human existence that addresses the physical, mental, and spiritual needs of study participants, which has contributed to improved treatment adherence and trend observed.^[46]

CONCLUSION

This study found an increase in MTB detection with Xpert MTB/RIF utilization. Ages >45 years have 2–4-fold increased risk of developing RR-TB and should be targeted for drug resistance prevention.