INT J TUBERC LUNG DIS 19(4):000–000 Q 2015 The Union http://dx.doi.org/10.5588/ijtld.14.0653
The impact of XpertW MTB/RIF in sparsely populated rural settings T. Van Den Handel,* K. H. Hampton,† I. Sanne,*‡ W. Stevens,§¶ R. Crous,# A. Van Rie† *Right to Care, Johannesburg, South Africa; †Department of Epidemiology, University of North Carolina Gillings School of Global Public Health, Chapel Hill, North Carolina, USA; ‡Clinical HIV Research Unit, Department of Medicine, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, §Department of Molecular Medicine and Hematology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, ¶National Health Laboratory Services, Johannesburg, #Rural Districts, Western Cape Department of Health, Cape Town, South Africa SUMMARY BACKGROUND:
The impact of implementing Xpertw MTB/RIF and the choice of instrument placement on patient care in sparsely populated areas with poor access to laboratory and radiology services have not yet been elucidated. M E T H O D S : Prospective evaluation of three diagnostic approaches in the Central Karoo, South Africa: smear microscopy as the initial diagnostic, with sputum processing at centralised laboratories, and Xpert as the initial diagnostic with instrument placement at facility level or centralised laboratory. R E S U LT S : Of 1449 individuals, 196 were diagnosed with TB. The proportion positive on initial testing was respectively 8%, 20% and 8% during the smear microscopy, decentralised Xpert and centralised Xpert
periods. The proportion of bacteriologically confirmed cases was respectively 88%, 99% and 91% during the smear microscopy, decentralised Xpert and centralised Xpert periods. The median time to treatment was respectively 11.5 (interquartile range [IQR] 6–24), 1 (IQR 0–2) and 6 days (IQR 2–9) during the smear microscopy, decentralised Xpert period and centralised Xpert periods. C O N C L U S I O N : Introducing Xpert as the initial diagnostic in areas with poor access to TB diagnostics increased the proportion of cases with bacteriological confirmation and reduced time to treatment initiation; however, point-of-care placement may have resulted in fewer people being evaluated for TB. K E Y W O R D S : impact; Xpert MTB/RIF; rural
TO REDUCE TUBERCULOSIS (TB) morbidity and mortality and limit onward transmission of Mycobacterium tuberculosis, TB control programmes aim for early detection of all cases, identification of drug resistance and timely initiation of appropriate treatment.1,2 A promising tool in the fight against TB is the Xpertw MTB/RIF (Cepheid, Sunnyvale, CA, USA) assay using the GeneXpert instrument, an automated diagnostic system that can rapidly detect M. tuberculosis and rifampicin (RMP) resistance with high sensitivity and specificity in sputum samples.3,4 As the system does not require specialised operator skills or bio-containment infrastructure, Xpert can be used at certain lower level health care facilities.5–7 The World Health Organization has endorsed Xpert as the initial diagnostic for TB in people suspected of having multidrug-resistant (MDR-) and human immunodeficiency virus (HIV) associated TB. In 2011, South Africa initiated national phased
implementation of Xpert for all individuals with presumptive TB. To facilitate rapid implementation and increase cost efficiency, the South African Department of Health (Pretoria, South Africa) decided to place all Xpert instruments at existing microscopy laboratory centres.8 While Xpert implementation has been shown to reduce the time to treatment initiation in urban settings,9–11 few data exist on the impact of Xpert in rural areas where people have less access to health care or laboratory facilities and are thus at higher risk of prolonged delay in diagnosis and treatment.1 We aimed to determine the impact of Xpert implementation on the time to anti-tuberculosis treatment initiation and the proportion of TB cases that were bacteriologically confirmed in a sparsely populated rural area in South Africa with poor access to diagnostic services and a prevalence of MDR-TB
Correspondence to: Theo Van Den Handel, Right to Care, Western Cape, P O Box 54 Plettenberg Bay, Western Cape 6600, South Africa. Tel: (þ27) 44 533 3067. Fax: (þ27) 44 533 3286. e-mail:
[email protected]; vdhandel@iafrica. com Article submitted 10 October 2014. Final version accepted 3 December 2014.
//titan/production/j/jtld/live_jobs/jtld-19-04/jtld-19-04-01/layouts/jtld-19-04-01.3d 29 January 2015 2:18 pm Allen Press, Inc. 0653
Page 1
2
The International Journal of Tuberculosis and Lung Disease
Figure 1 A) Map of South Africa with the Central Karoo district municipality highlighted and B) map of Central Karoo displaying the locations of health care and laboratory facilities. POC ¼ point of care.
and HIV-associated TB that would be considered low for South Africa, but intermediate globally.
STUDY POPULATION AND METHODS Setting The Central Karoo District (Karoo) of the Western Cape Province in South Africa is a rural, sparsely populated area covering 38 854 km2 —about the size of Switzerland— with a total population of 71 011 and population density of 1.8 inhabitants/km2, significantly lower than the average population density in the rest of South Africa (41.7 inhabitants/ km2).12 Almost half (48%) of the Karoo population lives in the capital city of Beaufort West; the remainder reside in towns with 5000–7000 inhabitants (Laingsburg, Murraysburg, Prince Albert), small towns of 1500–3000 inhabitants (Leeu Gamka, Nelspoort, Merweville) or small communities and
isolated farms (Figure 1, Table 1). The terrain is semidesert, and economic activity consists mostly of sheep and game farming, with little industry. Within the Western Cape Province, the Karoo has the highest percentage of deaths due to TB (8.4%) despite having the lowest HIV prevalence among TB cases (11.4%), the highest poverty (32.5%) and unemployment rates (30.8%), the lowest human development index (0.60), and the lowest number of health care facilities.12,13 Medical services are provided by 4 hospitals and 8 primary care clinics that are linked to 3 satellite clinics, 8 mobile clinics and 1 community day centre (Table 1, Figure 1). The only National Health Service Laboratory (NHLS) facility for TB diagnosis is located at the Beaufort West Hospital. All other facilities courier samples at least once a week to an NHLS laboratory located 60–160 km away (Table 1). Radiology services are available daily at the Beaufort
//titan/production/j/jtld/live_jobs/jtld-19-04/jtld-19-04-01/layouts/jtld-19-04-01.3d 29 January 2015 2:18 pm Allen Press, Inc. 0653
Page 2
Impact of Xpert in rural settings
3
Table 1 Characteristics of the health care facilities serving the population (n ¼ 71 011) of the Central Karoo, Western Cape Province, South Africa
Local municipality Beaufort West
Town
Population
Health care facilities
Beaufort West
34 085
1 district hospital; 3 primary care clinics; 2 mobile clinics; 1 community day centre 1 TB hospital; 1 primary care clinic; 1 mobile clinic 1 satellite clinic; 1 mobile clinic 1 subdistrict hospital; 1 primary care clinic; 1 mobile clinic 1 subdistrict hospital; 1 primary care clinic; 1 mobile clinic 1 satellite clinic 1 subdistrict hospital; 1 primary care clinic; 1 mobile clinic 1 primary care clinic; 1 mobile clinic 1 satellite clinic
Nelspoort
1 699
Merweville
1 592
Murraysburg*
5 069
Laingsburg
Laingsburg*
5 667
Prince Albert
Matjiesfontein* Prince Albert*
422 7 054
Leeu Gamka*
2 717
Klaarstroom*
584
Radiology services (number of days/week)
Laboratory courier (number of days/week)
NHLS laboratory facility
Distance to laboratory km
5
5
NHLS laboratory
,10
0
5
Beaufort West
55
0
1
Beaufort West
53
1
4
Beaufort West
160
1
3
Worcester
159
0 1
1 5
Worcester Oudtshoorn
139 113
0
3
Beaufort West
75
0
5
Oudtshoorn
60
* Facilities included in the analysis. NHLS ¼ National Health Service Laboratory; TB ¼ tuberculosis.
West Hospital, but only once weekly at the smaller hospitals. Study design We evaluated TB diagnosis and treatment initiation at six of the nine towns/communities in the Karoo (Table 1) during three distinct periods. Between April and October 2011 (the smear microscopy period), all sputum samples were sent for smear microscopy to the assigned NHLS laboratory. In October 2011, we placed a single one-module Xpert instrument in a safe, secure space at hospitals located in Laingsburg, Murraysburg and Prince Albert (the decentralised Xpert period), and the three nurses in charge of TB care were trained in performing the assay. Between November 2011 and April 2012, sputum samples were assessed using Xpert at point of care for individuals presenting at the hospitals in Laingsburg, Murraysburg and Prince Albert, or were transported to these hospitals for patients presenting at other care facilities (Table 1). Between July and December 2013, once all study cartridges had been used, sputum samples were again sent to the assigned NHLS laboratories, now for Xpert processing under the national roll-out programme (centralised Xpert period). Clients were screened for TB symptoms (cough 72 weeks, weight loss, night sweats) according to the South African Department of Health Guidelines.14 During the smear microscopy period, two sputum samples were collected per patient and transported by courier to the assigned NHLS laboratory for assess-
ment using fluorescent smear microscopy. In people with a history of anti-tuberculosis treatment, liquid culture and drug susceptibility testing (DST) were also requested. When Xpert was the initial diagnostic test, a single sputum sample was collected. If RMP resistance was detected on Xpert, a second sputum specimen was collected for culture and DST. During all three study periods, individuals without bacteriological confirmation of TB were prescribed antibiotics and advised to return after completion of the antibiotics if symptoms persisted, and were referred for assessment using chest X-ray (CXR) if clinically indicated. Individuals diagnosed with TB were managed according to Department of Health guidelines.14 Data collection, definitions and analysis Patient demographic (sex, age at first sputum collection) and clinical information (diagnostic testing dates and results, information about anti-tuberculosis treatment and HIV status) were collected from the TB suspect register, TB treatment register, HIV counselling and testing register, laboratory sputum register and Xpert instrument software output. A presumptive TB case was defined as an individual who presented with symptoms or signs suggestive of TB and who provided one or more sputum samples. A bacteriologically confirmed TB case was defined as an individual with a positive result on smear microscopy, culture or Xpert, independent of treatment initiation.15 A clinically diagnosed TB case was defined as an individual who did not fulfil the criteria for
//titan/production/j/jtld/live_jobs/jtld-19-04/jtld-19-04-01/layouts/jtld-19-04-01.3d 29 January 2015 2:18 pm Allen Press, Inc. 0653
Page 3
4
The International Journal of Tuberculosis and Lung Disease
bacteriological confirmation but was diagnosed with active TB by a health care worker (HCW) and initiated a full course of anti-tuberculosis treatment. Time to treatment initiation was defined as the time between the date of first sputum sample collection and the date of anti-tuberculosis treatment initiation. We compared time to treatment and proportion of cases with bacteriological confirmation between the three study periods: smear microscopy, decentralised Xpert and centralised Xpert. Data from the first weeks after placement of Xpert at the health care facilities were excluded to allow health care staff to adjust to the new TB diagnostic procedures. The length of this adjustment period differed by facility, and was determined by the time during which a mix of Xpert and smear microscopy was performed. The period between June 2012 and May 2013 was excluded due to an inconsistent supply of Xpert cartridges, after which a permanent stock was available at all three hospitals. Cartridge supply was dependent on donor funding, which then came to an end. We included the same number of clinic days for each of the three study periods of interest: 165 days for Laingsburg and Prince Albert and 180 days for Murraysburg. We used standard descriptive statistics and measured the association among variables using CochranMantel-Haenszel statistics, Pearson’s v2 test and Fisher’s exact tests. All analyses were performed using SAS, version 9.3 (Statistical Analysis System Institute, Cary, NC, USA). Maps were produced with ESRI ArcGIS v10.0 (Environmental Systems Research Institute, Redlands, CA, USA). Ethics approval Ethics approval was obtained from the University of the Witwatersrand Human Research Ethics Committee (HREC number: M110369), and the Institutional Review Board of the University of North Carolina at Chapel Hill, NC, USA. The requirement for written informed consent was waived. Study approval was also sought from the District and Provincial management, hospital directors and HIV/acquired immunedeficiency syndrome/sexually transmitted infection/ TB coordinators.
RESULTS Assessment of individuals with presumptive TB A total of 1449 individuals were assessed for TB over 1530 observation days, i.e., 510 observation days in each of the smear microscopy, decentralised Xpert and centralised Xpert periods, corresponding to 0.95 individuals per day in all facilities combined. Of the 1449 people assessed for TB, 584 (40%) were assessed during the smear microscopy period, 375 (26%) during the decentralised Xpert period, and 490 (34%) during the centralised Xpert period (Table 2).
During all three periods, just over half (53–54%) were male and .90% were adults (age .15 years). Tuberculosis case finding The proportion of positive tests among all initial tests (microscopy or Xpert) was 8% during the smear microscopy period, 20% during the decentralised Xpert period and 8% during the centralised Xpert period (Table 2). A total of 198 cases of TB were diagnosed: 75 during the smear microscopy period, 78 during the decentralised Xpert period and 45 during the centralised Xpert period. During the smear microscopy period, 88% (66/75) of cases were bacteriologically confirmed: most (48/66, 73%) using smear microscopy and some (18/66, 27%) using culture. Others were diagnosed solely on clinical presentation (7/75, 9%) or on CXR (2/75, 3%). During the decentralised Xpert period, 99% (77/78) of cases were bacteriologically confirmed (76 using Xpert, 1 using culture), only one case was diagnosed by CXR, and none clinically. During the centralised Xpert period, 91% (41/45) of cases were bacteriologically confirmed (38 using Xpert and 3 using culture); 2 cases were diagnosed by CXR and 2 clinically. Of 99 children evaluated for TB, 3 (3%) were diagnosed with active TB: 2 were microbiologically confirmed (1 by smear microscopy and 1 by Xpert) and 1 Xpertnegative child was diagnosed based on a suggestive CXR. Burden of HIV-associated TB and MDR-TB Of the 198 TB cases, 137 (69%) were tested for HIV. Among those tested, the HIV prevalence was similar during the three study periods (32%, 27% and 23%, respectively, P . 0.30 for all pairwise comparisons), lower than the estimated 65% HIV prevalence among TB patients in South Africa.16 Only 2 of the 198 cases were RMP-resistant (both detected using Xpert), of which one was also resistant to isoniazid, giving an MDR-TB prevalence of 0.5% (95% confidence interval [CI] 0.1–2.8). TB treatment initiation The proportion of bacteriologically confirmed TB cases who initiated treatment was lowest (89%) during the smear microscopy period and similar (99% or 100%) during the two Xpert periods (Table 2). Five of the seven individuals who failed to start treatment during the smear microscopy period had scanty acid-fast bacilli on a single sputum sample, one was smear-negative, culture-positive after 23 days and one was lost to follow-up. The median time to treatment was 11.5 days (interquartile range [IQR] 6–24) during the smear microscopy period: 8 days (IQR 4–13) in smearpositive individuals and 34 days (IQR 30–40) in smear-negative, culture-positive individuals (Figure
//titan/production/j/jtld/live_jobs/jtld-19-04/jtld-19-04-01/layouts/jtld-19-04-01.3d 29 January 2015 2:18 pm Allen Press, Inc. 0653
Page 4
Impact of Xpert in rural settings
5
Table 2 Characteristics of 1449 individuals evaluated using microbiological methods for presumptive TB and 196 TB cases diagnosed at health care facilities in Central Karoo, Western Cape Province, South Africa, stratified by type of initial assay (smear microscopy or XpertW MTB/RIF) and location of Xpert system (decentralised at health care facilities or centralised at laboratories) 1 Smear microscopy period n (%)
2 Decentralised Xpert period n (%)
3 Centralised Xpert period n (%)
Observation days
510 (33)
510 (33)
510 (33)
Individuals assessed for TB
584 (40)
375 (26)
490 (34)
Sex* Male Female
312 (54) 268 (46)
202 (54) 170 (46)
258 (53) 227 (47)
Age, years* ,5 5–14 715
27 (5) 25 (4) 528 (91)
1 (0) 10 (3) 361 (97)
13 (3) 23 (5) 449 (93)
48 (8) 75
76 (20) 78
38 (8) 45
18 (32) 38 (78)
15 (27) 41 (73)
10 (23) 33 (77)
76 (97) 1 (1) 1 (1)
38 3 2 2
Test positivity rate (smear or Xpert) Number of TB diagnoses made HIV status among TB cases† Positive Negative Basis of TB diagnosis Smear-positive Smear-negative, culture-positive Xpert-positive Xpert-negative, culture positive X-ray-positive Clinical symptoms only
1 vs. 2 P value
2 vs. 3 P value
1 vs. 3 P value
0.878
0.748
0.846
,0.001
0.005
0.235
,0.001
,0.001
0.780
0.535
0.689
0.332
48 (64) 18 (24) 2 (3) 7 (9
(84) (7) (4) (4)
Bacteriologically confirmed cases
66 (88)
77 (99)
41 (91)
0.008
0.059
0.765
Bacteriologically confirmed cases started on treatment
59 (89)
76 (99)
41 (100)
0.024
1.000
0.042
* Missing in 12 individuals (4 in smear microscopy period, 3 in facility-Xpert period and 5 in laboratory-Xpert period). † Percentage is expressed among those diagnosed with TB tested for HIV; HIV status was missing in respectively 19 (25%), 22 (28%) and 2 (4%) TB cases in the smear microscopy, decentralised Xpert and centralised Xpert periods. TB ¼ tuberculosis; HIV ¼ human immunodeficiency virus.
2). The median time to treatment dropped to 1 day (IQR 0–2) during the decentralised Xpert period and increased again to 6 days (IQR 2–9) during the centralised Xpert period.
DISCUSSION In the sparsely populated communities of the Karoo, access to the Xpert assay and choice of placement of the Xpert instrument influenced the diagnostic decision making of HCWs and time to treatment initiation. The proportion of people receiving a bacteriologically confirmed diagnosis of TB increased by 11% (from 88% to 99%), and HCWs were less likely to make a clinical or radiological diagnosis of TB when centralised smear microscopy was replaced by decentralised Xpert. When sputum samples were no longer processed at the facility level but sent to centralised laboratories for Xpert, HCWs reverted to making more clinical and radiological diagnoses, reducing the proportion of patients with bacteriologically confirmed cases from 99% to 91%. The total number of cases detected did not increase following the introduction of Xpert, but a higher proportion of cases diagnosed initiated anti-
tuberculosis treatment (91% vs. 99% or 100%), and treatment was initiated sooner following sputum collection. The switch from smear microscopy to Xpert at centralised NHLS laboratories reduced the median time to treatment by 5.5 days, from 11.5 to 6 days. The median time to treatment of 6 days during the centralised Xpert period was similar to the 8-day delay for smear-positive patients during the smear microscopy period, suggesting that the faster treatment initiation following the introduction of Xpert is attributable to the higher sensitivity of Xpert compared to smear microscopy, thereby reducing reliance on the lengthy culture process. Placement of Xpert at facility level during the decentralised Xpert period enabled same-day treatment for most patients, and reduced the time to treatment by a median of 5 days (from 6 days to 1 day) compared to the centralised Xpert period. Enabling same-day treatment initiation is important for economically deprived individuals in the rural Karoo who often have to travel long distances to reach the health care facilities. The observed increase in bacteriological confirmation and reduced time to treatment are similar to observations in urban settings in sub-Saharan Africa.10 The lack of increase in case finding is in contrast
//titan/production/j/jtld/live_jobs/jtld-19-04/jtld-19-04-01/layouts/jtld-19-04-01.3d 29 January 2015 2:18 pm Allen Press, Inc. 0653
Page 5
6
The International Journal of Tuberculosis and Lung Disease
may limit the generalisability of the findings to other settings.
CONCLUSIONS
Figure 2 Time to treatment initiation from first sputum collection among 196 tuberculosis cases diagnosed during three study periods: smear microscopy, decentralised Xpert and centralised Xpert period. Full horizontal line represents the median, box represents the 25th to 75th percentiles, short horizontal bars represent the 5th and 95th percentiles, open diamond represents the mean and open circles the outliers. * One outlier (treatment started at 94 days) not shown on graph. IQR ¼ interquartile range.
to model predictions,8 but is in line with two recent studies that found little or no impact of Xpert on case finding compared to smear microscopy,9,17 due, in part, to high rates of empiric treatment. In contrast to other findings, the lower treatment initiation rate we observed during the smear microscopy period was not due to loss to follow-up during the diagnostic process,18 but was primarily due to HCWs not treating those individuals with a single scanty positive smear microscopy result. Placement of the diagnostic assay also influenced the positivity rate of the initial test: 8% of samples sent to centralised laboratories for smear microscopy were positive, compared to 20% of samples processed by Xpert at facility level, and 8% of samples processed using Xpert in centralised laboratories. With good case detection, approximately 10% of sputum samples should be smear-positive.19 These findings suggest that HCWs restricted the use of Xpert when it was performed at facility level. Clinic staff attributed this finding to stricter adherence to TB screening criteria because of the perceived high cost of Xpert. An additional factor may have been the extra time required by HCWs to perform the Xpert assay. While our results are one of the first to demonstrate the impact of Xpert in rural areas, they must be interpreted in the light of its limitations. Notwithstanding the 6-month duration of each of the three study periods and the inclusion of 91% of the Central Karoo population outside of the capital city, the low population density resulted in small numbers of TB cases diagnosed in each of the three study periods. The unique characteristics of the Karoo population
Our results show a substantial benefit of introducing Xpert as the initial diagnostic in areas with poor access to TB diagnostics, even when the burden of HIV-associated and MDR-TB is relatively low. The limited investment of the decentralised placement of three single one-module Xpert instruments and the training of one nurse per health care facility increased the proportion of patients receiving a bacteriological confirmation of TB and reduced the time to treatment initiation, but may have resulted in fewer people being evaluated for TB. In addition to targeting individuals suspected of HIV-associated or MDR-TB as currently recommended, prioritising Xpert to areas with poor access to TB diagnostics could expand the patient impact of Xpert implementation. Acknowledgements The authors would like to thank the Western Cape Department of Health and especially H Schumann (District Director) and A Jooste (District HAST Director) and their team: J Nel, F Fass, A Adams and R Koorts. A Swartz, M Poolman and J Arendse at the Provincial office for their support; I Noordien and L Scott from the National Health Laboratory Services (Johannesburg, South Africa) for their input and advice; L Page-Shipp for support; and M Karsten and the Right to Care Western Cape team for their assistance. Funding for this study was provided by United States Agency for International Development/President’s Emergency Plan for AIDS Relief under Agreement No. 674-A-00-08-0007-00 to Right to Care. Conflicts of interest: none declared.
References 1 Storla D G, Yimer S, Bjune G A. A systematic review of delay in the diagnosis and treatment of tuberculosis. BMC Public Health 2008; 8: 15. 2 Weyer K, Mirzayev F, Migliori G B, et al. Rapid molecular TB diagnosis: evidence, policy making and global implementation of Xpert MTB/RIF. Eur Respir J 2013; 42: 252–271. 3 Boehme C C, Nabeta P, Hillemann D, et al. Rapid molecular detection of tuberculosis and rifampin resistance. N Engl J Med 2010; 363: 1005–1015. 4 Steingart K R, Schiller I, Horne D J, Pai M, Boehme C C, Dendukuri N. Xpertw MTB/RIF assay for pulmonary tuberculosis and rifampicin resistance in adults. Cochrane Database Syst Rev 2014; 1: CD009593. 5 Van Rie A, Page-Shipp L, Scott L, Sanne I, Stevens W. Xpertw MTB/RIF for point-of-care diagnosis of TB in high-HIV burden, resource-limited countries: hype or hope? Expert Rev Mol Diagn 2010; 10: 937–946. 6 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: 229–237. 7 Raizada N, Sachdeva K S, Sreenivas A, et al. Feasibility of decentralised deployment of Xpert MTB/RIF test at lower level of health system in India. PLOS ONE 2014; 9: e89301. 8 Meyer-Rath G, Schnippel K, Long L, et al. The impact and cost of scaling up GeneXpert MTB/RIF in South Africa. PLOS ONE 2012; 7: e36966.
//titan/production/j/jtld/live_jobs/jtld-19-04/jtld-19-04-01/layouts/jtld-19-04-01.3d 29 January 2015 2:18 pm Allen Press, Inc. 0653
Page 6
Impact of Xpert in rural settings
9 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, controlled trial. Lancet 2014; 383: 424–435. 10 Boehme C C, Nicol M P, Nabeta P, et al. Feasibility, diagnostic accuracy, and effectiveness of decentralised use of the Xpert MTB/RIF test for diagnosis of tuberculosis and multidrug resistance: a multicentre implementation study. Lancet 2011; 377: 1495–1505. 11 Hanrahan C F, Selibas K, Deery C B, et al. Time to treatment and patient outcomes among TB suspects screened by a single point-of-care xpert MTB/RIF at a primary care clinic in Johannesburg, South Africa. PLOS ONE 2013; 8: e65421. 12 Regional Development profile, Central Karoo District. Cape Town, South Africa: Western Cape Government, Provincial Treasury, 2012. 13 Statistics South Africa. Mortality and causes of death in South Africa, 2010: findings from death notification. Pretoria, South Africa: Statistics South Africa, 2013.
7
14 South African National Department of Health. National tuberculosis guidelines. Pretoria, South Africa: National Department of Health, 2009. 15 World Health Organization. Definitions and reporting framework for tuberculosis: 2013 revision. Geneva, Switzerland: WHO, 2013. 16 World Health Organization. Global tuberculosis report, 2013. WHO/HTM/TB/2013.11. Geneva, Switzerland: WHO, 2013. 17 Churchyard G. Effect of Xpert MTB/RIF on early mortality in adults with suspected TB: a pragmatic randomized trial. Boston, MA, USA: 21st Conference on Retroviruses and Opportunistic Infections (CROI 2014), 3–6 March 2014. [Oral Abstract 95]. 18 Botha E, den Boon S, Lawrence K A, et al. From suspect to patient: tuberculosis diagnosis and treatment initiation in health facilities in South Africa. Int J Tuberc Lung Dis 2008; 12: 936–941. 19 de Kantor I N, Kim S, Frieden T R, et al. Laboratory services in tuberculosis control Part 2: Microscopy. Geneva, Switzerland: WHO, 1998.
//titan/production/j/jtld/live_jobs/jtld-19-04/jtld-19-04-01/layouts/jtld-19-04-01.3d 29 January 2015 2:18 pm Allen Press, Inc. 0653
Page 7
Impact of Xpert in rural settings
i
RESUME
L’impact de la mise en œuvre du Xpertw MTB/RIF et du choix de l’emplacement des appareils de laboratoire sur la prise en charge du patient dans des r´egions peu peupl´ees avec un acc`es m´ediocre aux services de laboratoire et de radiologie n’a pas encore e´ t´e e´ lucid´e. M E´ T H O D E S : Evaluation prospective de trois approches diagnostiques dans la r´egion de Central Karoo, Afrique du Sud : microscopie de frottis pour le diagnostic initial avec traitement des crachats dans des laboratoires centralis´es et Xpert pour le diagnostic initial avec une affectation des appareils au niveau des structures de soins ou d’un laboratoire central. R E´ S U LT A T S : Au total, 196 cas de tuberculose (TB) ont e´ t´e diagnostiqu´es chez 1449 individus. La proportion de cas positifs parmi les tests initiaux a e´ t´e de 8%, 20% et 8% lors de la p´eriode de microscopie de frottis, de Xpert CONTEXTE :
d´ecentralis´e et de Xpert centralis´e, respectivement. La proportion de cas confirm´es bact´eriologiquement a e´ t´e de 88%, 99% et 91% pendant la p´eriode de microscopie de frottis, de Xpert d´ecentralis´e et de Xpert centralis´e, respectivement. Le d´elai m´edian de traitement a e´ t´e de 11,5 jours (IQR 6–24), 1 jour (IQR 0–2) et 6 jours (IQR 2–9) pendant la p´eriode de microscopie de frottis, de Xpert d´ecentralis´e et de Xpert centralis´e, respectivement. C O N C L U S I O N : Introduire le Xpert comme diagnostic initial dans des re´ gions ayant un acce` s limite´ au diagnostic de TB a accru la proportion de cas avec confirmation bact´eriologique et amen´e une r´eduction du d´elai de mise en route du traitement anti-tuberculeux, mais la localisation des tests aux points de service peut avoir diminu´e le nombre de personnes b´en´eficiant d’un diagnostic. RESUMEN
M A R C O D E R E F E R E N C I A: No se ha estudiado la repercusion ´ de la introduccion ´ de la prueba Xpertw MTB/RIF y de la ubicacion ´ del dispositivo sobre la atencion ´ al paciente en las zonas de baja densidad de poblacion ´ y con escaso acceso a los servicios de laboratorio y radiolog´ıa. M E T O D O S: En el distrito Central Karoo de Sura´frica se llev o´ a cabo una evaluaci on ´ prospectiva de tres estrategias diagnosticas, ´ a saber: la baciloscopia como prueba diagnostica ´ inicial con procesamiento del esputo en laboratorios centralizados, la prueba Xpert como examen inicial y ubicacion ´ del dispositivo en el centro de atencion ´ o en un laboratorio centralizado. R E S U L T A D O S: Se diagnosticaron 196 casos de tuberculosis (TB) en 1449 personas. La proporcion ´ de pruebas iniciales positivas fue 8% durante el per´ıodo de la baciloscopia, 20% con la prueba Xpert descentralizada y 8% durante el per´ıodo de la prueba
Xpert en un laboratorio central. La proporcion ´ de casos confirmados bacteriologicamente ´ fue 88% durante el periodo de la baciloscopia, 99% con la prueba Xpert descentralizada y 91% con la prueba Xpert centralizada. La mediana del lapso hasta comenzar el tratamiento fue 11,5 d´ıas (intervalo intercuartil [IQR] 6–24) con la baciloscopia, 1 d´ıa (IQR 0–2) durante el per´ıodo de la prueba Xpert perif´erica y 6 d´ıas (IQR 2–9) con la prueba Xpert en el laboratorio central. C O N C L U S I O N: La introduccion ´ de la prueba Xpert como el examen diagnostico ´ inicial en zonas con escaso acceso a los m´etodos diagnosticos ´ de la TB aument o´ la proporci on ´ de casos confirmados bacteriologicamente ´ y disminuyo´ el lapso hasta el comienzo del tratamiento antituberculoso; sin embargo, la ubicacion ´ del dispositivo en el punto de atencion ´ tal vez disminuyo´ el numero ´ de personas en quienes se investigo´ el diagnostico ´ de TB.
//titan/production/j/jtld/live_jobs/jtld-19-04/jtld-19-04-01/layouts/jtld-19-04-01.3d 29 January 2015 2:18 pm Allen Press, Inc. 0653
Page 8
Queries for jtld-19-04-01 This manuscript/text has been typeset from the submitted material. Please check this proof carefully to make sure there have been no font conversion errors or inadvertent formatting errors. Allen Press.
//titan/production/j/jtld/live_jobs/jtld-19-04/jtld-19-04-01/layouts/jtld-19-04-01q.3d Thursday, 29 January 2015 2:18 pm Allen Press, Inc. Page 1
