Hrev_master [page 10] [Healthcare in Low-resource Settings 2015; 3:5011] Incremental detection of pulmonary tuberculosis among presumptive patients by GeneXpert MTB/RIF® over fluorescent microscopy in Mwanza, Tanzania: an operational study Jeremiah Seni,1 Benson R. Kidenya,1 Mercy Anga,1 Anthony Kapesa,1 John R. Meda,2 Richard Mutakyawa,3,4 Zahra H. Mkomwa,4 Fidelis Marcel,3 John M. Changalucha,5 Stephen E. Mshana1 1Department of Microbiology and Immunology, Catholic University of Health and Allied Sciences Bugando, Mwanza 2Department of Internal Medicine, University of Dodoma 3Sekou Toure Regional Referral Hospital, Mwanza 4PATH - Tanzania, Dar es Salaam 5National Institute for Medical Research, Mwanza Medical Research Centre, Tanzania Abstract Laboratory confirmation among presump- tive tuberculosis (PTB) patients is pivotal in ensuring prompt management. Limited infor- mation exists in Tanzania regarding the per- formance of GeneXpert MTB/RIF® in compari- son with conventional methods. An operational study was conducted involving 806 PTB patients at Sekou Toure Hospital in Mwanza, Tanzania from June to November 2013. Patients’ information was obtained and their respective sputum samples analyzed by light- emitting diode fluorescent microscopy (LED FM) and GeneXpert MTB/RIF®. The mean age of study participants was 39.6±16.0 years, with males accounting for 50.5%. The majority of patients (97.5%) were new cases. The propor- tions of PTB patients confirmed by LED FM and GeneXpert MTB/RIF® were 14.1% (114/806) and 23.7% (191/806) respectively, resulting into a 9.6% incremental detection rate by GeneXpert MTB/RIF® over LED FM. The detec- tion rate among HIV positive individuals was also higher [23.6% (63/267) vs 14.2% (38/267), respectively], with an incremental detection of 9.4%. The incremental detection of PTB by GeneXpert MTB/RIF® over LED FM calls for expansion of its use to increase detection of smear negative PTB among people living with HIV. Introduction The escalating burden of tuberculosis (TB) in Tanzania in the midst of high prevalence of HIV/AIDS poses a negative social and econom- ic impact in this developing country which is ranked 22nd among countries accounting for 80% of the global burden of TB.1-3 To avert continuous transmission, morbidity and mortality attributable to TB, laboratory confirmation among presumptive pulmonary tuberculosis (PTB) patients is pivotal in ensur- ing prompt management.2,4 Ziehl-Neelsen (ZN)-based light microscopy which is the main stay and universally available diagnostic tech- nique in Tanzania and other developing coun- tries has long been shown to have low perform- ance.5,6 In the light of this, the World Health Organization (WHO) has recommended scal- ing up the use of light-emitting diode fluores- cent microscopy (LED FM) which is on average 10% more sensitive in detection of TB com- pared to the conventional ZN-based light microscopy using culture as a gold standard.5,7,8 This notwithstanding, LED FM coverage is still low in developing countries.2 To address the low performance of microscopy-based detec- tion methods for TB, a number of molecular based diagnostic methods have been validated by WHO to increase coverage and enhance timely detection of PTB patients,8-10 but their utility is unevenly appreciated across coun- tries mainly due to the installation and run- ning costs as well as lack of expertize.10,11 Recently, WHO endorsed a new rapid molec- ular test called GeneXpert MTB/RIF® (Cepheid, Sunnyvale, CA, USA).9 The dual function of the machine in simultaneously diagnosing TB and identifying resistance to one of the core first line anti-TB drug, rifampicin along with its high sensitivity and specificity, has revolution- ized the diagnosis of TB globally.9,12-15 The per- formance of GeneXpert MTB/RIF® has been shown to be better compared to LED FM in both smear positive and negative people living with HIV (PLWH), though variability exists depend- ing on the population involved.16-18 The rifampin resistance has been shown to vary in different countries from 0% in Mbeya (Tanzania), 10% in Harare (Zimbabwe) to as high as 35.1% in Moldova.16,19 In response to WHO call to scale up the utilization of this new diagnostic, the Ministry of Health in the United Republic of Tanzania, through the National Tuberculosis and Leprosy Control Program (NTLP)3,9 and other developmental partners, has cordially rolled out the GeneXpert MTB/RIF® machines to various regions. Apparently the target groups are smear nega- tive PLWH, PTB patient who recently contacted multidrug resistant tuberculosis (MDR) patient and children.3 In Tanzania, Mwanza region is second to Dar Es Salaam in terms of TB case notification rates emphasizing the need to have reliable diagnostic methods in place.3 Despite this, lim- ited information exists in this region regard- ing the performance of the recently introduced GeneXpert MTB/RIF® in comparison with LED FM for the diagnosis of TB. Furthermore, the magnitude of rifampicin (RIF) resistance remains to be explored in this setting. Therefore, the present study aimed at deter- mining the incremental detection of TB among PTB patients by GeneXpert MTB/RIF® and LED FM at Sekou Toure Regional Referral Hospital (SRRH) in Mwanza, Tanzania so as to offer baseline information crucial for future assess- ment of the diagnostic performance of the facility as well as the utility of the new tech- nique in this local setting. Healthcare in Low-resource Settings 2015; volume 3:5011 Correspondence: Jeremiah Seni, Department of Microbiology and Immunology, Catholic University of Health and Allied Sciences, P.O. Box 1464, Bugando, Mwanza, Tanzania. Tel: +255.78.4593000 - Fax: +255.28.2502678. E-mail: senijj80@gmail.com Key words: Tuberculosis detection; GeneXpert MTB/RIF®; Mwanza; Tanzania. Contributions: JS, BRK, MA and SEM conceived and designed the study; MA and FM carried out the laboratory procedures; JS, BRK, MA, AK and JRM analyzed data; JS wrote the first draft of the manuscript; AK, JRM, RM, ZHM, JC and SEM crit- ically reviewed the manuscript. All authors have read and approved the final draft of the manu- script. Conflict of interest: the authors declare no poten- tial conflict of interest. Acknowledgements: the authors are sincerely thankful to the patients who participated in the study, SRRH administration for allowing conduc- tion of this study. Mr. Othman Sade and other lab- oratory staffs working in the TB section at SRRH for their technical support. The GeneXpert MTB/RIF® was generously donated and is being maintained PATH Tanzania under USAID TB TO 2015 Funds. Part of this work was presented at the 6th CUHAS Scientific Graduation Symposium: Abstract Book, November 2014, Mwanza. Tanzania. Received for publication: 17 January 2015. Revision received: 21 March 2015. Accepted for publication: 26 March 2015. This work is licensed under a Creative Commons Attribution 3.0 License (by-nc 3.0). ©Copyright J. Seni et al., 2015 Licensee PAGEPress, Italy Healthcare in Low-resource Settings 2015; 3:5011 doi:10.4081/hls.2015.5011 No n c om me rci al us e o nly [Healthcare in Low-resource Settings 2015; 3:5011] [page 11] Materials and Methods Study design and area This was an operational prospective labora- tory based study carried out at SRRH in Mwanza, Tanzania from June 2013 to November 2013 involving 1946 PTB patients submitting their sputum for analysis at SRRH. Of these, 806 (41.4%) had dual results (i.e. LED FM and GeneXpert MTB/RIF® results) ful- filling the inclusion criteria, and 1140 (58.6%) patients were excluded for various reasons (Figure 1). Sample collection, processing and data analysis Sputum samples were collected from PTB patients following the NTLP Guidelines,1 and analyzed based on the standard operating pro- cedures by LED FM and GeneXpert MTB/RIF®.8,9,15,20 For comparison purposes of the two diagnostic techniques, one sputum sample per patient was used. In case the sam- ple was negative requiring the second sputum sample as per NTLP Guideline,1 the latter was analyzed to guide patient’s management but not used for the index study. Patients’ information was obtained from lab- oratory request forms and the TB registry book. Analysis was done using STATA software ver- sion 11 (College Station, TX, USA) according to the objectives of the study. Continuous vari- ables were described as mean (±standard devi- ation). Categorical variables were described as proportions (percentages) and were analyzed to compare the distribution of PTB positive and negative patients with variables. Study clearance and ethical consid- erations The study was approved by the joint Bugando Medical Centre and Catholic University of Health and Allied Sciences Institutional Review Board. Permission to conduct the study was obtained from SRRH medical officer in charge, TB coordinator and laboratory manager. All patients’ information was kept confidential and anonymous using study codes. Presumptive patients found to have PTB were treated in their respective treatment units basing on the NTLP guidelines1 and those with RIF resistance were referred to Kibong’oto National Tuberculosis Hospital for confirmation and further expertize management. Results We involved 806 PTB patients in this study with the mean age (±standard deviation) of 39.6±16.0 years (age range 1-96 years); males accounted for 50.5% (407/806). Majority of patients (97.5%) were new cases and were residing within Mwanza City, 81.7% (658/806). The proportion of PTB patients confirmed to have PTB disease by either FM or GeneXpert MTB/RIF® was 24.8% (200/806) (Table 1). Of these, 14.1% (114/806) and 23.7% (191/806) were detected by FM and GeneXpert MTB/RIF® respectively. This resulted into 9.6% incremen- tal detection rate by GeneXpert MTB/RIF® over LED FM (Figure 1 and Table 1). The detection rate of GeneXpert MTB/RIF® was higher compared to LED FM in both chil- dren (≤17 years) [8.3% (6/72) vs 4.2% (4/72)] and adults [25.2% (185/734) vs 15.1% (111/734)] respectively resulting into the incremental detections of 4.1% and 10.1% for children and adults respectively. The detection rate among HIV positive individuals was also higher using GeneXpert MTB/RIF® compared to FM [23.6% (63/267) vs 14.2% (38/267) respectively], with an incremental detection of 9.4%. Moreover, the GeneXpert MTB/RIF® detected 12.4% (86/692) and 12.2% (28/229) among all smear negative irrespective of HIV serostatus and smear negative HIV positive PTB patients were respectively. The RIF resistance was found in 2 (1.1%) patients, 5 (2.6%) had indeterminate resist- ance, whereas in 184 (96.3%) there was no rifampin resistance detected. Of 200 PTB posi- tive patients, majority were found to be in the age group of more than 18 years (96.5%), males (60%), residents of Mwanza city (69.4%), new cases (94.5%) and HIV positive (86.8%) (Table 2). Discussion The low performance of sputum smear microscopy in developing countries with high Article Table 1. Diagnostic performance of GeneXpert MTB/RIF® vs light emitting diode fluo- rescent microscopy. LED FM GeneXpert MTB/RIF® Total MTB detected MTB not detected AFB detected 105 9 114 AFB not detected 86 606 692 Total 191 615 806 LED FM, light emitting diode fluorescent microscopy; MTB, Mycobacterium tuberculosis; RIF, rifampicin; AFB, acid fast bacilli. Figure 1. Flow chart showing series of events in the recruitment procedures and results. LED FM, Light emitting diode fluorescent microscopy; PTB, pulmonary tuberculosis; MTB, Mycobacterium tuberculosis; RIF, rifampicin; SRRH, Sekou Toure Regional Referral Hospital. No n c om me rci al us e o nly [page 12] [Healthcare in Low-resource Settings 2015; 3:5011] TB burden has been widely documented and if unchecked, it can result into uninterrupted transmission of this deadly infectious disease.2,5,6 Despite a number of new techno- logical advancement on the diagnosis of TB, the local evaluations of their performance remain a challenge in most developing coun- tries.10,11,13,21 The incremental detection of 9.6% among PTB patients at SRRH by GeneXpert MTB/RIF® over LED FM in the present study along with the 23% from a review involving 8880 participants in 21 studies,12 8.0% among children in Uganda,17 and 9.7% in a recent multicenter, randomized controlled trial involv- ing South Africa, Zimbabwe, Zambia and Tanzania,16 emphasizes the utility of GeneXpert MTB/RIF® over microscopy in the diagnosis of TB patients. But the cost-related challenges for the universal introduction of GeneXpert MTB/RIF® in many health facilities in developing countries reiterate the need to continue strengthening the pre-existing microscopy-based TB diagnostic methods, so that the newer technique remains reserved to risky groups like smear negative PLWH, PTB patients who recently contacted MDR, and chil- dren.3 The incremental detection of TB among smear negative PLWH in this and other stud- ies17,18,21,22 further justifies its utility in this risky group as recommended by the new NTLP guidelines.3 The use of GeneXpert MTB/RIF® to detect RIF resistance as a surrogate marker of MDR has been suggested in many studies, with concordance ranging from 88 to 100%.14,23,24 In the light of these, RIF resistance in the present study (2.2%) is higher than 0.86 (4/464) and 0.17% (2/1167) from a study in Mwanza and National survey in Tanzania respectively25,26 but lower than 3.5 to 7.3% in different African countries.19 Interestingly, no RIF resistance has been detected in three stud- ies from Mbeya, Tanzania.16,21,27 The finding of RIF resistance in Mwanza region which is sec- ond to Dar Es Salaam in terms of TB case noti- fication calls for strengthening of surveillance system in this region to enable timely detec- tion of patients with RIF resistant and MDR TB, thereby interrupting further transmission by provision of prompt management. Based on the nature of works and likelihood of exposure, the preponderance of males and city dwelling residents to be infected with PTB in this study is also similar to other reports.3,28 The high proportion of PTB patients to be co-infected with HIV in the present study relates to anoth- er study.17 These findings are also supported by other studies which have shown association of development of active TB with HIV/AIDS, smoking, co-morbidity such as diabetes melli- tus, indoor air pollution and young age.2,29 Limitations The culture method which is a gold standard for laboratory diagnosis of TB is not done at SRRH. Thus, this operational study did not compare the performance of GeneXpert MTB/RIF® and LED FM with culture. Also, the impact of other predictor variables on diagnos- tic performance such as CD4+ count was not evaluated. Conclusions There is an approximately 10% incremental detection of TB among PTB patients by GeneXpert MTB/RIF® compared to LED FM, with more detection also among smear nega- tive PLWH who are apparently targeted by NTLP to be among beneficiaries of this new technology. Therefore, we recommend the expansion of its use to increase detection of PTB among smear negative PLWH at SRRH and other settings in the Lake Victoria zone. Evaluation of GeneXpert MTB/RIF® perform- ance among people with extra pulmonary TB and the impact of various predictor variables on this diagnostic assay will be of interest to further delineate its utility in this setting. References 1. Ministry of Health and Social Welfare, United Republic of Tanzania. Manual of the national tuberculosis and leprosy pro- gramme in Tanzania. Dar es Salaam, Tanzania: Ministry of Health and Social Welfare; 2006. 2. WHO. Global tuberculosis control: WHO report 2011. Geneva, Switzerland: World Health Organization; 2011. Available from: http://whqlibdoc.who.int/publications/2011 /9789241564380_eng.pdf 3. Ministry of Health and Social Welfare, United Republic of Tanzania. Manual for the management of tuberculosis and lep- rosy. National tuberculosis and leprosy programme. Dar es Salaam, Tanzania: Ministry of Health and Social Welfare; 2013. 4. Zumla A, Raviglione M, Hafner R, von Reyn CF. Tuberculosis. New Engl J Med 2013;368:745-55. 5. Steingart KR, Henry M, Ng V, et al. Fluorescence. Conventional sputum smear microscopy for tuberculosis: a systematic review. Lancet Infect Dis 2006;6:570-81. 6. Seni J, Kidenya BR, Obassy E, et al. Low sputum smear positive tuberculosis among pulmonary tuberculosis suspects in a tertiary hospital in Mwanza, Tanzania. Tanzania J Health Res 2012;14:1-9. 7. Cattamanchi A, Davis JL, Worodria W, et al. Sensitivity and specificity of fluores- cence microscopy for diagnosing pul- monary tuberculosis in a high HIV preva- lence setting. Int J Tuberc Lung D 2009; 13:1130-6. 8. WHO. Fluorescent light-emitting diode (LED) microscopy for diagnosis of tuber- culosis: policy statement. Geneva, Switzerland: World Health Organization; 2011. Available from: http://whqlibdoc.who. Article Table 2. Distribution of pulmonary tuberculosis positive and negative patients with variables. Variables PTB patients (total=806) Positive (total=200)*n (%) Negative (total=606) n (%) Mean age (years) 39.4±14.0° 39.7±16.7° Age groups (years) ≤8 4 (2.0) 17 (2.8) 9-17 3 (1.5) 48 (7.9) ≥18 193 (96.5) 541 (89.3) Sex Female 80 (40.0) 319 (52.6) Males 120 (60.0) 287 (47.4) Residence Mwanza City 138 (69.4) 520 (85.8) Outside Mwanza City 61 (30.6) 86 (14.2) Treatment category New cases 189 (94.5) 597 (98.5) Follow up 11 (5.5) 9 (1.5) HIV serostatus# Positive 66 (86.8) 201 (87.0) Negative 10 (13.2) 30 (13.0) PTB, presumptive tuberculosis. *Diagnosed by either LED FM or GeneXpert MTB/RIF®; °continuous variable; #only 307 patients knew HIV serostatus. No n c om me rci al us e o nly [Healthcare in Low-resource Settings 2015; 3:5011] [page 13] int/publications/2011/9789241501613_eng .pdf 9. WHO. Policy statement: automated real- time nucleic acid amplification technology for rapid and simultaneous detection of tuberculosis and rifampicin resistance: Xpert MTB/RIF system. Geneva, Switzerland: World Health Organization; 2011. Available from: http://whqlibdoc.who.int/publications/2011 /9789241501545_eng.pdf 10. Parsons LM, Somoskovi A, Gutierrez C, et al. Laboratory diagnosis of tuberculosis in resource-poor countries: challenges and opportunities. Clin Microbiol Rev 2011;24:314-50. 11. Pantoja A, Fitzpatrick C, Vassall A, et al. Xpert MTB/RIF for diagnosis of tuberculo- sis and drug-resistant tuberculosis: a cost and affordability analysis. Eur Respir J 2011;42:708-20. 12. Steingart KR, Schiller I, Horne DJ, et al. Xpert(R) MTB/RIF assay for pulmonary tuberculosis and rifampicin resistance in adults. Cochrane Db Syst Rev 2014:CD009593. 13. Lawn SD, Nicol MP. Xpert(R) MTB/RIF assay: development, evaluation and imple- mentation of a new rapid molecular diag- nostic for tuberculosis and rifampicin resistance. Future Microbiol 2011;6:1067- 82. 14. Boehme CC, Nabeta P, Hillemann D, et al. Rapid molecular detection of tuberculosis and rifampin resistance. New Engl J Med 2010;363:1005-15. 15. Helb D, Jones M, Story E, et al. Rapid detection of Mycobacterium tuberculosis and rifampin resistance by use of on- demand, near-patient technology. J Clin Microbiol 2010;48:229-37. 16. Theron G, Zijenah L, Chanda D, et al. Feasibility, accuracy, and clinical effect of point-of-care Xpert MTB/RIF testing for tuberculosis in primary-care settings in Africa: a multicentre, randomised, con- trolled trial. Lancet 2014;383:424-35. 17. Sekadde MP, Wobudeya E, Joloba ML, et al. Evaluation of the Xpert MTB/RIF test for the diagnosis of childhood pulmonary tuberculosis in Uganda: a cross-sectional diagnostic study. BMC Infect Dis 2013;13:133. 18. Ssengooba W, Nakiyingi L, Armstrong DT, et al. Clinical utility of a novel molecular assay in various combination strategies with existing methods for diagnosis of HIV-related tuberculosis in Uganda. PLoS One 2014;9:e107595. 19. Creswell J, Codlin AJ, Andre E, et al. Results from early programmatic imple- mentation of Xpert MTB/RIF testing in nine countries. BMC Infect Dis 2014;14:2. 20. Lumb R, Van Deun A, Bastlan I, Fitz-Gerald M. Laboratory diagnosis of tuberculosis by sputum microscopy. Adelaide, Australia: SA Pathology; 2010. Available from: http://www.who.int/tb/laboratory/tb-spu- tum-microscopy-handbook.pdf 21. Rachow A, Zumla A, Heinrich N, et al. Rapid and accurate detection of Mycobacterium tuberculosis in sputum samples by Cepheid Xpert MTB/RIF assay: a clinical validation study. PLoS One 2011;6:e20458. 22. Lawn SD, Brooks SV, Kranzer K, et al. Screening for HIV-associated tuberculosis and rifampicin resistance before anti- retroviral therapy using the Xpert MTB/RIF assay: a prospective study. PLoS Med 2011;8:e1001067. 23. Kidenya BR, Webster LE, Behan S, et al. Epidemiology and genetic diversity of mul- tidrug-resistant tuberculosis in East Africa. Tuberculosis 2014;94:1-7. 24. Blakemore R, Story E, Helb D, et al. Evaluation of the analytical performance of the Xpert MTB/RIF assay. J Clin Microbiol 2011;48:2495-501. 25. Range N, Friis H, Mfaume S, et al. Anti- tuberculosis drug resistance pattern among pulmonary tuberculosis patients with or without HIV infection in Mwanza, Tanzania. Tanzania J Health Res 2012;14:1-9. 26. Chonde TM, Basra D, Mfinanga SG, et al. National anti-tuberculosis drug resistance study in Tanzania. Int J Tuberc Lung D 2010;14:967-72. 27. Ntinginya EN, Squire SB, Millington KA, et al. Performance of the Xpert(R) MTB/RIF assay in an active case-finding strategy: a pilot study from Tanzania. Int J Tuberc Lung D 2012;16:1468-70. 28. Austin JF, Dick JM, Zwarenstein M. Gender disparity amongst TB suspects and new TB patients according to data record- ed at the South African Institute of Medical Research laboratory for the Western Cape Region of South Africa. Int J Tuberc Lung D 2004;8:435-9. 29. Narasimhan P, Wood J, Macintyre CR, Mathai D. Risk factors for tuberculosis. Pulm Med 2013; 2013:828939. Article No n c om me rci al us e o nly