Background: Bamako, Mali, has a well-structured hierarchy for tuberculosis (TB) case management. However, in recent years Bamako has been faced with an emerging threat from multidrug resistant TB (MDR-TB). Here we present insights into the genomic epidemiology of TB and the evolutionary mechanisms driving the emergence of MDR-TB in Bamako. Methods & Materials: Isolates recovered from tuberculosis patients from local reference centers and the University Teaching Hospital at Point G, in Bamako, Mali between 2006 and 2012 (n = 208), were tested for antimicrobial susceptibility at the MRC Unit The Gambia. A subset of 76 isolates were analysed using whole genome sequencing. A time dated phylogenetic tree was reconstructed using BEAST. Lineage and resistance conferring mutations were inferred using PhyResSe. Results: Patients included 21 females and 55 males aged between 3 to 78 years, among whom 12(16%) were infected by MDR-TB. Most patients 61(80%) were new cases and among 15 retreatment cases 9(60%) were MDR-TB. The phylogeny was reconstructed from 8508 variant core genome sites. The dominant lineage was the Euro-American super lineage, lineage 4. Within lineage 4, the Cameroon genotype was the most prevalent genotype (n = 20, 26%) followed by the Ghana genotype (n = 16, 21%). Cameroon genotype isolates diverged from a common recent ancestor ∼161 years ago to form three clusters, one of which emerged ∼22 years ago and is likely to be involved in on-going transmission. Seven Ghana genotype isolates were MDR-TB representing over half all MDR-TB in this dataset (7/12). Ghana genotype isolates were more likely to cause MDR-TB than other genotypes after controlling for treatment status (OR = 5.6, p-value = 0.043). The MDR-TB Ghana genotype isolates formed a clade that diverged approximately 30 years ago, in which the katGSer315Thr mutation was conserved. Other Euro-American genotypes included the six LAM, two H37Rv-like and one Uganda. Four patients were infected with closely related Beijing strains and five patients were infected with non-MDR Mycobacterium africanum2. Conclusion: The Cameroon and Ghana genotypes are endemic to West Africa. Their association with on-going spread and MDR-TB in Bamako respectively is a major concern. The rise of Beijing genotype in Bamako is worrying given its high transmissibility and virulence.

Background: The Gambia is a high TB burden country but childhood TB is under recognised because of difficulties with diagnosis and low priority accorded to it by the TB Program. Methods: We implemented contact tracing to identify co-prevalent tuberculosis among child contacts of adults with smear positive tuberculosis in the Greater Banjul Area of the country where over 70% of all TB cases are diagnosed. Symptoms screening and tuberculin skin test (TST) were performed in the community. All symptomatic and/or TST positive child contacts were brought to the dedicated childhood TB clinic at the MRC where physical examination, chest x ray and sputum examination were performed. Results: Altogether, 4042 child contacts were enrolled in the study following screening in the community; 1772 were aged less than 5 years whilst 2270 were aged 5 years and above. Of the 4042, 82.6% (3339/4042) were diagnosed as TB exposed but not infected, 15.8% (639/4042) were latently infected and 1.6% (64/4042) had co-prevalent TB disease. Of the 64 TB cases, 19 (29.7%) were bacteriologically confirmed while 45 (70.5%) were clinically diagnosed. Fifty (78.1%) of the 64 TB cases were diagnosed among child contacts in the same household as the adult index case whilst 14 (21.9%) were outside the household. Of 630 asymptomatic but TST positive child contacts evaluated in the clinic, 27(4.3%) had coprevalent TB disease, comprising 42.8% of all TB cases diagnosed. Symptom screening alone would have detected only 57.8% of co-prevalent cases. Conclusion: This study showed that in our community setting, contact tracing detected coprevalent TB cases among child contacts both within and outside the household of the adult index case. Contact tracing restricted to symptom screening and to the index cases’ households would have missed nearly half of all co-prevalent TB disease in child contacts.

SETTING:Greater Banjul Area of the Gambia.OBJECTIVES:To identify co-prevalent tuberculosis (TB) among child contacts of adults with smear-positive TB.DESIGN:Child contacts aged <15 years in the immediate household and compound were prospectively enrolled and evaluated for TB disease using screening questionnaires and the tuberculin skin test (TST). Symptomatic and/or TST-positive (10 mm) contacts were further investigated.RESULTS:Of 4042 child contacts who underwent symptom screening and TST, 3339 (82.6%) were diagnosed as TB-exposed but not infected, 639 (15.8%) were latently infected and 64 (1.6%) had co-prevalent TB. Of the 64 TB cases, 50 (78.1%) were from within the immediate household of the index case, and 14 (21.9%) from within the same compound. Of the 27 asymptomatic but TST-positive children diagnosed with TB, 7 were microbiologically confirmed. The median age of the TB cases was 4.4 years (interquartile range 1.9-6.9); 53.1% were aged <5 years. Of the 4042 child contacts, 206 (5%) slept in the same bed as the index case; 28.1% of all TB cases occurred in this group. Symptom screening alone would have detected only 57.8% of the co-prevalent cases.CONCLUSION:In our community setting, if contact tracing is restricted to symptom screening and immediate households only, nearly half of all co-prevalent TB disease in child contacts would be missed.

Background: An outbreak of pneumococcal meningitis among non-infant children and adults occurred in the Brong-Ahafo region of Ghana between December 2015 and April 2016 despite the recent nationwide implementation of a vaccination programme for infants with the 13-valent pneumococcal conjugate vaccine (PCV13).Methods: Cerebrospinal fluid (CSF) specimens were collected from patients with suspected meningitis in the Brong-Ahafo region. CSF specimens were subjected to Gram staining, culture and rapid antigen testing. Quantitative PCR was performed to identify pneumococcus, meningococcus and Haemophilus influenzae. Latex agglutination and molecular serotyping were performed on samples. Antibiogram and whole genome sequencing were performed on pneumococcal isolates.Results: Eight hundred eighty six patients were reported with suspected meningitis in the Brong-Ahafo region during the period of the outbreak. In the epicenter district, the prevalence was as high as 363 suspected cases per 100,000 people. Over 95 % of suspected cases occurred in non-infant children and adults, with a median age of 20 years. Bacterial meningitis was confirmed in just under a quarter of CSF specimens tested. Pneumococcus, meningococcus and Group B Streptococcus accounted for 77 %, 22 % and 1 % of confirmed cases respectively. The vast majority of serotyped pneumococci (80 %) belonged to serotype 1. Most of the pneumococcal isolates tested were susceptible to a broad range of antibiotics, with the exception of two pneumococcal serotype 1 strains that were resistant to both penicillin and trimethoprim-sulfamethoxazole. All sequenced pneumococcal serotype 1 strains belong to Sequence Type (ST) 303 in the hypervirulent ST217 clonal complex.Conclusion: The occurrence of a pneumococcal serotype 1 meningitis outbreak three years after the introduction of PCV13 is alarming and calls for strengthening of meningitis surveillance and a re-evaluation of the current vaccination programme in high risk countries.

Background Although Drug resistance tuberculosis is not a new phenomenon, Mali remains one of the “blank” countries without systematic data. Methods Between 2006 and 2014, we enrolled pulmonary TB patients from local TB diagnostics centers and a university referral hospital in several observational cohort studies. These consecutive patients had first line drug susceptibility testing (DST) performed on their isolates. A subset of MDR was subsequently tested for second line drug resistance. Results A total of 1186 mycobacterial cultures were performed on samples from 522 patients, including 1105 sputa and 81 blood samples, yielding one or more Mycobacterium tuberculosis complex (Mtbc) positive cultures for 343 patients. Phenotypic DST was performed on 337 (98.3%) unique Mtbc isolates, of which 127 (37.7%) were resistant to at least one drug, including 75 (22.3%) with multidrug resistance (MDR). The overall prevalence of MDR-TB was 3.4% among new patients and 66.3% among retreatment patients. Second line DST was available for 38 (50.7%) of MDR patients and seven (18.4%) had resistance to either fluoroquinolones or second-line injectable drugs. Conclusion The drug resistance levels, including MDR, found in this study are relatively high, likely related to the selected referral population. While worrisome, the numbers remained stable over the study period. These findings prompt a nationwide drug resistance survey, as well as continuous surveillance of all retreatment patients, which will provide more accurate results on countrywide drug resistance rates and ensure that MDR patients access appropriate second line treatment.

OBJECTIVE/BACKGROUND:To evaluate the Kudoh swab method for improving laboratory diagnosis of tuberculosis (TB) in Gambia. METHODS:A total of 75 sputa (50 smear positive and 25 smear negative) were examined. Sputum samples were collected from leftover routine samples from the Medical Research Council Unit, Gambia TB Diagnostic Laboratory. The samples were processed using the standard N-acetyl-l-cysteine-NaOH (NALC-NaOH) methods currently used and Kudoh swab method. These were cultured on standard Lowenstein Jensen (LJ) and Modified Ogawa media, respectively, and incubated aerobically at 36±1°C for mycobacterial growth. To determine if the decontamination and culture methods compared could equally detect the Mycobacterium tuberculosis complex (MTBC) highly commonly isolated in Gambia, spoligotyping was done. RESULTS:In total, 72% (54/75) of MTBC were recovered by both LJ and Modified Ogawa methods. The LJ method recovered 52% (39/75) and Modified Ogawa recovered 56% (42/75) of the MTBC, respectively. Spoligotyping showed Euro-American 35% (19/54), Indo-Oceanic 35% (19/54), Mycobacterium africanum (West African type 2) 26% (14/54), Beijing 2% (1/54), and M. africanum (West African type 1) 2% (1/54). CONCLUSION:The Kudoh method is simpler and cheaper than the NALC-NaOH method. There was no significant difference in recovery between the methods. The Kudoh method is ideal in overburdened TB laboratories with poor resources in developing countries. The predominant lineages were Euro-American and Indo-Oceanic, followed by M. africanum (West African type 2).

Objective/Background: To determine the culture yield and time to detection of mycobacterial growth between samples decontaminated using 0.7% chlorhexidine and sodium hydroxide–N-acetyl-l-cysteine (NaOH–NALC) and cultured on the Löwenstein–Jensen (LJ) medium. We also aimed to determine the contamination rate between the 0.7% chlorhexidine and NaOH–NALC decontamination methods. Methods: The study was carried out on 68 sputa samples (42 smear positives and 26 smear negatives). Of these 68 samples, 46 were collected from men and 26 from women with an approximate average age of 27 years. All the sputum samples were decontaminated using the standard NaOH–NALC and 0.7% chlorhexidine methods. The concentrates were cultured in parallel on LJ media in which reading of the slope for mycobacterial growth was obtained daily for the first 2 weeks and then weekly until week 8. The mycobacterial recovery rate, time to detection, and contamination rate were then compared. Results: The overall recovery rate of mycobacterial growth on samples treated with both decontamination methods inoculated on LJ media is 51.5% (35/68). Specifically, mycobacterial growth rates on samples treated with 0.7% chlorhexidine and standard NaOH–NALC on LJ media were 61.8% (42/68) and 54.4% (37/68), respectively. However, the growth of Mycobacterium tuberculosis complex was faster on samples treated with 0.7% chlorhexidine than those treated with NaOH–NALC (average, 32 ± 5 days vs. 33 ± 5.2 days, respectively). The contamination rate on samples treated with 0.7% chlorhexidine was 1.5% (1/68), whereas on those treated with NaOH–NALC, the rate was 4.4% (3/68). Conclusion: The 0.7% chlorhexidine decontamination method is rapid and has less contamination rate in terms of mycobacterial recovery compared with the standard NaOH–NALC method. Therefore, the 0.7% chlorhexidine decontamination method would be an ideal alternative option for decontamination of sputum samples and recovery/isolation of M. tuberculosis in resource-poor countries.

SETTINGGreater Banjul Area, The Gambia.OBJECTIVETo conduct a pragmatic evaluation of the Xpert(®) MTB/RIF assay in the diagnosis of tuberculosis (TB) among child contacts.DESIGNIn this prospective study, one induced sputum sample was obtained from TB contacts aged <15 years and tested using fluorescent microscopy, culture and Xpert. The diagnostic accuracy of the microbiological tests was evaluated against culture and 'all TB diagnosis and treatment' as separate reference standards.RESULTSUsing culture as a reference standard, Xpert was positive for Mycobacterium tuberculosis in 6/14 culture-positive and 6/473 culture-negative children, giving a sensitivity and specificity of respectively 42.9% (95%CI 17.7-71.1) and 98.7% (95%CI 97.2-99.5). With 'all TB diagnosis and treatment' as a composite reference standard, combined Xpert and culture tests were positive for M. tuberculosis in 20/62 children with TB disease (32.3%, 95%CI 20.9-45.3), which was comparable to the yield from microscopy, culture and Xpert combined (33.9%, 95%CI 22.3-47.0), but significantly higher than individual yields from each test.CONCLUSIONThe sensitivity of Xpert is low in actively traced child contacts, but a combination of Xpert and mycobacterial culture has incremental benefits for the bacteriological confirmation of TB disease.

To determine the rates of drug-resistant tuberculosis in The Gambia, Mycobacterium tuberculosis isolates obtained from 225 patients during a nationwide survey were tested against isoniazid, rifampicin, ethambutol and streptomycin using the resistance ratio method. Only nine (4%) of the patients had strains that were resistant to one or more drugs. None of the patients with drug-resistant M. tuberculosis had previously been treated for tuberculosis. Drug-resistant tuberculosis is, as yet, not common in The Gambia. Periodic surveys for drug-resistant tuberculosis are recommended to monitor changes that may emerge over time.