Risk factors for loss to follow-up, transfer or death among people living with HIV on their first antiretroviral therapy regimen in Mali.

OBJECTIVES: Risk factors for loss to follow-up (LTFU) were assessed for people living with HIV (PLHIV) at various reference out-patient clinics (expertise level II) and hospitals (expertise level III) in Mali. METHODS: HIV-1-positive adults starting antiretroviral therapy (ART) in 2006-2013 were eligible for inclusion. Risk factors for LTFU, defined as no visit in the 6 months preceding the last database update, were assessed with the Cox model, taking into account the competing risks of transfer and death. Potential risk factors at the start of ART were demographic and socioeconomic variables, World Health Organization (WHO) stage, CD4 count, period of ART initiation, type of ART, region of care, expertise level and distance from home. RESULTS: We included 9821 PLHIV, 33% of whom were male, starting ART at nine out-patient clinics and seven hospitals [five and two in the capital Bamako and four and five in the 'regions' (i.e. districts outside the capital), respectively] with a median (interquartile range) CD4 count of 153 (56-270) cells/µL. Five-year cumulative incidences of LTFU, transfer and death were 35.2, 9.7 and 6.7%, respectively. People followed at Bamako hospitals > 5 km from home, at regional hospitals or at regional out-patient clinics < 5 km from home were at higher risk of LTFU than people followed at Bamako out-patient clinics, whereas people followed at regional out-patient clinics 5-50 km away from home were at lower risk for LTFU. Deaths were less frequent at hospitals, whether in Bamako or in the regions, than at Bamako out-patient clinics, and more frequent at regional out-patient clinics. CONCLUSIONS: Expertise level and distance to care were associated with LTFU. Stigmatization may play a role for PLHIV living close to the centres in the regions.

High level of HIV-1 resistance in patients failing long-term first-line antiretroviral therapy in Mali.

OBJECTIVES: In resource-limited settings, few data are available on virological failure after long-term first-line antiretroviral therapy. This study characterized the genotypic resistance patterns at the time of failure after at least 36 months of a first-line regimen in Mali, West Africa. METHODS: Plasma samples from 84 patients who were receiving first-line antiretroviral treatment and with an HIV-1 RNA viral load (VL) >1000 copies/mL were analysed. Genotypic resistance testing was performed and HIV-1 drug resistance was interpreted according to the latest version of the National Agency for HIV and Hepatitis Research algorithm. RESULTS: At the time of resistance testing, patients had been treated for a median of 60 months (IQR 36-132 months) and had a median CD4 cell count of 292 cells/mm(3) (IQR 6-1319 cells/mm(3)), a median HIV-1 RNA level of 28266 copies/mL (IQR 1000-2 93 495 copies/mL) and a median genotypic susceptibility score of 1 (IQR 1-4). The prevalence of nucleoside reverse transcriptase inhibitor (NRTI) and non-nucleoside reverse transcriptase inhibitor (NNRTI) resistance mutations was 78% and 82%, respectively. Viruses were resistant to at least one drug in 92% of cases. Although etravirine and rilpivirine were not used in the first-line regimens, viruses were resistant to etravirine in 34% of cases and to rilpivirine in 49% of cases. The treatment duration, median number of NRTI and NNRTI mutations and some reverse transcriptase mutations (T215Y/F/N, L210W, L74I, M41L and H221Y) were associated with the VL at virological failure. CONCLUSIONS: This study demonstrated a high level of resistance to NRTIs and NNRTIs, compromising second-generation NNRTIs, for patients who stayed on long-term first-line regimens. It is crucial to expand the accessibility of virological testing in resource-limited settings to limit the expansion of resistance and preserve second-line treatment efficacy.

Genetic barrier to the development of resistance to rilpivirine and etravirine between HIV-1 subtypes CRF02_AG and B.

OBJECTIVES: It has been demonstrated for some drugs that the genetic barrier, defined as the number of genetic transitions and/or transversions needed to produce a resistance mutation, can differ between HIV-1 subtypes. We aimed to assess differences in the genetic barrier for the evolution of resistance to the second-generation non-nucleoside reverse transcriptase inhibitors etravirine and rilpivirine in subtypes B and CRF02_AG in antiretroviral-naive patients. METHODS: An analysis was undertaken of 25 substitutions associated with etravirine and rilpivirine resistance at 12 amino acid positions in 267 nucleotide sequences (136 HIV-1 B and 131 HIV-1 CRF02_AG subtypes) of the reverse transcriptase gene. RESULTS: The majority (7/12) of amino acid positions studied were conserved between the two HIV-1 subtypes, leading to a similar genetic barrier. Different predominant codons between the subtypes were observed in 5/12 positions (90, 98, 179, 181 and 227), with an effect on the calculated genetic barrier only at the V179D and V179F codons (2.5 versus 3.5 for V179D, and 2.5 versus 5 for V179F, respectively, for subtype B versus subtype CRF02_AG). CONCLUSIONS: The majority of amino acids involved in etravirine and rilpivirine resistance showed a high degree of conservation of the predominant codon between the B and CRF02_AG subtypes. For rilpivirine, the genetic barrier was the same between the two subtypes. Nevertheless, subtype CRF02_AG showed a higher genetic barrier to acquiring mutations V179D and V179F (mutations associated with resistance to etravirine) compared with subtype B, suggesting that it would be more difficult to produce resistance to etravirine in the CRF02_AG subtype than the B subtype.

[Socio-demographic, clinical, and therapeutic aspects of adolescents treated and followed for HIV infection at the Hospital Gabriel Toure paediatric center]. / Profil epidemio-clinique et therapeutique du VIH de l'adolescent au departement de pediatrie du chu Gabriel Toure.

HIV infection is a chronic infectious disease requiring long-term management and regular follow-up of patients. OBJECTIVES: The aims of this study was to describe the socio-demographic, clinical, biological and therapeutic aspects of adolescents treated and followed for HIV infection at the Hospital Gabriel Toure paediatric center. PATIENTS AND METHODS: From 01/01/2001 to 31/12/2017, the medical records of children followed for HIV infection until adolescence were analyzed. It was a descriptive and analytical retrospective study. RESULTS: One thousand five hundred and fourteen patients received antiretroviral treatment and 587 were still in follow-up on 31 December 2017, including 393 adolescents (sex-ratio = 1.2). The median age was 14.25 years and 55.1% of children had lost at least one parent. HIV serology was positive among mothers in 61.7% of cases (n=342), and 63% of them were on ARVs. Sixty-eight per cent of children were WHO Stage III or IV at the time of ART initiation. The median age at onset of ART was 53 months (26-96 months). The combination of 2 nucleotide reverse transcriptase inhibitors (NRTIs) with a non-nucleotide reverse transcriptase inhibitor (NNRTI) was used in 89% of patients. The median CD4 count before ARV treatment was 438/mm3. The average duration of follow-up under treatment was 9.8 ± 3.4 years. Fifty-one percent of adolescents had undetectable viral load. There was a correlation between the initiation of a second line of treatment and treatment failure (p<0.001). CONCLUSION: The adherence of adolescents to ARV treatment requires the implementation of innovative strategies to improve the therapeutic success rate.

L'infection à VIH est une maladie chronique infectieuse nécessitant une prise en charge longue et un suivi régulier des patients. OBJECTIFS: L'objectif de ce travail était de décrire les aspects socio-démographiques, clinico-biologiques et thérapeutiques du VIH chez l'adolescentau centre d'excellence pédiatrique de prise en charge du CHU Gabriel Toure. PATIENTS ET MÉTHODES: Il s'agissait d'une étude transversale avec recueil rétrospectif de données, qui s'est déroulée du 01/01/2001 au 31/12/2017. C'était une étudetransversale à visée analytique portant sur les dossiers des adolescents d'au moins10 ans. RÉSULTATS: Trois cent quatre-vingt-treize (393) adolescents d'au moins10 ansont été inclus. La sérologie VIH était positive chez les mères dans 61,7% des cas (n=342), et 63% d'entre elles étaient sous ARV. Soixante-huit pour cent des enfants étaient classés stade III ou IV de l'OMS au moment de la mise sous TARV. L'association de 2 inhibiteurs nucléosidiques de la transcriptase inverse (INTI) à un inhibiteur non nucléotidique de la transcriptase inverse (INNTI) a été utilisée chez 89% des patients. Le taux de CD4 médian avant la mise sous traitement ARV était de 438/mm3 La durée moyenne de suivi sous traitement était de 9,8 ± 3,4 ans. Cinquante un pourcent (51%) des adolescents étaient en succès thérapeutique avec une charge virale indétectable (< 1000 copies/ml). Il y avait une corrélation entre l'instauration d'une seconde ligne de traitement et l'échec thérapeutique (p<0,001). CONCLUSION: L'adhésion des adolescents au traitement ARV nécessite la mise en place de stratégies innovantes permettant d'améliorer le taux de succès thérapeutique.

Second-line antiretroviral therapy failure and characterization of HIV-1 drug resistance patterns in children in Mali.

INTRODUCTION: In recent years, children born to HIV-infected mothers have been receiving antiretroviral treatment (ART) with limited or no virologic monitoring, which increases the likelihood of development and accumulation of drug resistance mutations, which itself may limit the effectiveness of future ART. The objective of this study was to evaluate the prevalence of resistance mutations in children infected with HIV-1 experiencing virological failure to second-line ART in the Pediatric Department of Gabriel Touré Hospital in Mali. METHODS: Children aged from 5 to 18 infected with HIV-1 on second-line antiretroviral therapy and whose viral load was greater than 1000 copies/mL after observance reinforcement were enrolled. The protease and reverse transcriptase genes were sequenced with ViroSeq®. The results were interpreted according to the last version of the Stanford algorithm in 2018. The study was approved by the Ethics Committee of the Faculty of Medicine and Dentistry, University of Sciences, Techniques and Technologies of Bamako (Mali). RESULTS: Of 216 children, 33 (15.3%) who had a viral load (VL)>1000 copies/mL in second line were recruited and included in the study. The median plasma viral load was 77,000 copies/mL [IQR (28,000-290,000)] and the median CD4 cell count was 310 cells/mm3 [IQR (152-412)]. The median age was 12 years; 48.5% of patients were treated with a combination of stavudine/lamivudine/nevirapine (Triomune®) for first-line treatment and 60.6% with abacavir/lamivudine/lopinavir/ritonavir for the second-line ART. The median treatment duration was 8.5 years [range, 3-13]. Of the 33 children whose treatment failed, the predominant HIV-1 subtype was CRF02_AG (66.7%). The prevalence of resistance to ART classes was 60.61% (20/33) to nucleoside reverse transcriptase inhibitors (NRTIs), 54.51% (18/33) to nonnucleoside reverse transcriptase inhibitors (NNRTIs), and 51.52% (17/33) to protease inhibitors (PIs). Of the patients studied, 90.9% were exposed to lopinavir/ritonavir (LPV/r) but only 15.2% (5/33) developed resistance to LPV/r. CONCLUSIONS: This study demonstrated that LPV/r remains active in most patients after second-line ART failure. In children whose second-line ART fails, particular attention should be paid to their ART and adherence history when considering the next treatment option.

Prevalence of extended-spectrum beta-lactamase phenotypes in enterobacteria isolated from blood cultures of patients at admission to the University Hospital of Bamako.

The aim of our study was to determine the frequency of extended-spectrum beta-lactamase (ESBL) phenotypes among the enterobacteria present in blood cultures of patients at admission to two university hospitals of Bamako (Mali). During a period of three months, we isolated enterobacteria from blood cultures from patients upon admission to the Point G and Gabriel Toure University Hospitals. The ESBL-positive enterobacteria were initially identified by API 20E strips and VITEK®2 and then confirmed in France by MALDI-TOF mass spectrometry at the Bichat Hospital bacteriology laboratory. Antibiotic susceptibility was determined by the diffusion method as recommended by EUCAST. The species isolated were K. pneumoniae (14/40, 35.0 %), E. coli (11/40, 27.5 %), and E. cloacae (9/40, 22.5 %); 21/34 (61.8 %) had an ESBL phenotype, including 10/14 (71.4 %) K. pneumoniae, 8/11 (72.7 %) E. coli, and 3/9 (33 3 %), E. cloacae. The ESBL strains of K. pneumoniae, E. coli, and E. cloacae were associated, respectively, with resistance to the following antibiotics: gentamicin (10/10, 100 %; 6/8, 75%; 2/3, 67%), amikacin (2/10, 20 %; 0/8, 0%; 0/3, 0%), ofloxacin (8/10, 80. %; 7/8, 87%; 3/3, 100%), cotrimoxazole (10/10, 100 %; 6/8, 75%; 3/3, 100%). Almost two thirds (61.8%) of the enterobacteria isolated from blood cultures produced extended-spectrum beta-lactamases. They retained regular sensitivity only to carbapenems and amikacin.

Prevalence of extended-spectrum beta-lactamase-producing Enterobacteriaceae isolated from blood cultures in Africa.

Extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae have been isolated from many regions of the world. Epidemiological studies are being conducted in Europe, North America, and Asia. No study has however been conducted in Africa to determine the prevalence and distribution of ESBLs on the continent. This literature review aimed at describing the prevalence of ESBL-producing Enterobacteriaceae isolated from blood cultures, as well as the ESBL genes involved at the international level. Our focus was mainly on Africa. We conducted a literature review on PubMed. Articles related to our study field and published between 1996 and 2014 were reviewed and entirely read for most of them, while we only focused on the abstracts of some other articles. Relevant articles to our study were then carefully reviewed and included in the review. The prevalence of ESBL-producing Enterobacteriaceae differs from one country to another. The results of our literature review however indicate that class A ESBLs prevail over the other types. We took into consideration articles focusing on various types of samples to assess the prevalence of ESBL-producing Enterobacteriaceae, but information on isolates from blood cultures is limited. The worldwide prevalence of ESBL-producing Enterobacteriaceae has increased over time. Evidence of ESBL-producing Enterobacteriaceae can be found in all regions of the world. Studies conducted in Africa mainly focused on the Northern and Eastern parts of the continent, while only rare studies were carried out in the rest of the continent.

[HIV/TB co-infection in rural settings of Benin: case of the Djougou-Ouake-Copargo sanitary district (north-west Benin)]. / Co-infection VIH/Tuberculose en milieu rural au Benin: cas de la zone sanitaire Djougou-Ouake-Copargo (nord-ouest Benin).

AIMS: we measured the burden of TB/HIV co-infection in a rural setting of Benin, and assessed the outcome of tuberculosis treatment at the end of the intensive phase of TB treatment. METHODS: This is a retrospective, cross-sectional, descriptive study, covering January 2006 to December 2011. RESULTS: A total of 256 patients were included, 67 (26.1%) were HIV +. A minority, 25% of co-infected HIV / PTB, had TB bacilli high density (+++) versus 45% of mono-infected (P = 0.005). The smear conversion was obtained in 96% of coinfected versus 93% in HIV- at the end of the intensive phase (P = 0.5). The cure rate was 86% and 93.1% respectively in co-infected and HIV-. A proportion of 13.5% of co-infected died versus 3% in HIV- (P = 0.005). 21% of co-infected with CD4 <200 died versus 3.6% of those with CD4> 200 (P = 0.041). CONCLUSION: This work underlines the high prevalence of HIV / TB co-infection in this region. Co-infected patients respond well to treatment, but their mortality is high when they are very immunocompromised.

BUTS: nous avons mesuré le fardeau que constitue la co-infection VIH/ tuberculose chez des tuberculeux en milieu rural au Bénin ; et évaluer l'issue du traitement antituberculeux à la fin de la phase intensive. MÉTHODES: Il s'agit d'une étude transversale rétrospective, descriptive couvrant Janvier 2006 à Décembre 2011. RÉSULTATS: Au total 256 patients ont été colligés, 67 (26,1%) étaient VIH+. Une proportion de 25 % des co-infectés VIH/TPM+ avaient une densité bacillaire à trois croix (+++) contre 45% des tuberculeux VIH- (P=0,005). La négativation de la bacilloscopie était obtenue chez 96% des patients co-infectés contre 93% chez les tuberculeux VIH- à la fin de la phase intensive (P=0,5). Le taux de guérison était respectivement de 86% et 93,1% chez les co-infectés et les non VIH. Une proportion de 13,5% des co-infectés étaient décédés contre 3% chez les VIH- (P=0,005). 21% des co-infectés ayant un CD4<200 étaient décédés contre 3,6% de ceux dont le CD4>200 (P=0,041). CONCLUSION: Ce travail souligne la forte prévalence de l'infection par le VIH chez les tuberculeux de cette région. Les co-infectés répondent bien au traitement, mais leur taux de mortalité est plus élevé, surtout s'ils sont très immunodéprimés.

[HIV/TB co-infection in rural settings of benin: case of Djougou-Ouake-Copargo district (north-west Benin]. / Co-infection VIH/Tuberculose en milieu rural au benin: cas de la zone sanitaire Djougou-Ouake-Copargo (Nord-ouest Benin).

AIMS: We measured the burden of HIV/tuberculosis (HIV/TB) co-infection in people infected by TB in rural settings of Benin, and assessed the outcome of TB treatment at the end of the intensive phase. METHODS: This is a retrospective, cross-sectional, descriptive study, covering January 2006 to December 2011. RESULTS: A total of 256 patients were gathered, 67 (26.1%) were HIV +. A proportion of 25% of co-infected HIV / PTB had TB bacilli high density (+++) versus 45% of mono-infected (P = 0.005). The smear conversion was obtained in 96% of coinfected versus 93% in HIV- at the end of the intensive phase (P = 0.5). The cure rate was 86% and 93.1% respectively in co-infected and HIV-. A proportion of 13.5% of co-infected died versus 3% in HIV- (P = 0.005). 21% of co-infected with CD4 <200 died versus 3.6% of those with CD4> 200 (P = 0.041). CONCLUSION: This work underlines the high prevalence of HIV / TB co-infection in this region. Co-infected patients respond well to treatment, but their mortality is high when they are very immunocopromissed.

BUTS: Nous avons mesuré le fardeau que constitue la co-infection VIH/ tuberculose chez des tuberculeux en milieu rural au Bénin; et évaluer l'issue du traitement antituberculeux à la fin de la phase intensive. MÉTHODES: Il s'agit d'une étude transversale rétrospective, descriptive couvrant Janvier 2006 à Décembre 2011. RÉSULTATS: Au total 256 patients ont été colligés, 67 (26,1%) étaient VIH+. Une proportion de 25 % des co-infectés VIH/TPM+ avaient une densité bacillaire à trois croix (+++) contre 45% des tuberculeux VIH- (P=0,005). La négativation de la bacilloscopie était obtenue chez 96% des patients co-infectés contre 93% chez les tuberculeux VIH- à la fin de la phase intensive (P=0,5). Le taux de guérison était respectivement de 86% et 93,1% chez les co-infectés et les non VIH. Une proportion de 13,5% des co-infectés étaient décédés contre 3% chez les VIH- (P=0,005). 21% des co-infectés ayant un CD4<200 étaient décédés contre 3,6% de ceux dont le CD4>200 (P=0,041). CONCLUSION: Ce travail souligne la forte prévalence de l'infection par le VIH chez les tuberculeux de cette région. Les co-infectés répondent bien au traitement, mais leur taux de mortalité est plus élevé, surtout s'ils sont très immunodéprimés.