Time for a revolution in tracking the HIV epidemic

In the past two decades, thanks to effective antiretroviral treatment (ART), enormous progress has been made in improving the health and life expectancy of HIV-infected individuals.1Wandeler G Johnson LF Egger M Trends in life expectancy of HIV-positive adults on antiretroviral therapy across the globe: comparisons with general population.Curr Opin HIV AIDS. 2016; (published online May 31.)https://doi.org/10.1097/COH.0000000000000298Crossref PubMed Scopus (214) Google Scholar The success of ART, combined with the benefits of treatment as prevention and pre-exposure prophylaxis on HIV transmission, have generated new hope to end the HIV/AIDS epidemics.2Vermund SH Hayes RJ Combination Prevention: New Hope for Stopping the Epidemic.Curr HIV/AIDS Rep. 2013; 10: 169-186Crossref PubMed Scopus (63) Google Scholar Translation of this hope into universal reality will need accelerated efforts to ensure rapid access to ART and to extend HIV prevention services to reach the most affected populations and regions. These advances will only be feasible with an accurate and detailed picture of the HIV pandemic, and especially of the trends in HIV incidence—one of the most crucial epidemiological indicators and the most challenging to estimate.3Brookmeyer R Measuring the HIV/AIDS epidemic: approaches and challenges.Epidemiol Rev. 2010; 32: 26-37Crossref PubMed Scopus (99) Google Scholar For many years, global estimates of the HIV burden have been produced by UNAIDS only, and have been largely based on HIV prevalence data and complex mathematical models.4Hallett TB Zaba B Stover J et al.Embracing different approaches to estimating HIV incidence, prevalence and mortality.AIDS. 2014; 28: S523-S532Crossref PubMed Scopus (13) Google Scholar In 2014, Christopher Murray and colleagues5Murray CJ Ortblad KF Guinovart C et al.Global, regional, and national incidence and mortality for HIV, tuberculosis, and malaria during 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013.Lancet. 2014; 384: 1005-1070Summary Full Text Full Text PDF PubMed Scopus (694) Google Scholar provided an alternative assessment of HIV incidence, prevalence, and mortality for the Global Burden of Disease Study (GBD) 2013. The investigators used complex mathematical models and specified underlying assumptions to estimate HIV incidence and prevalence in each country, mainly from the number of deaths caused by HIV recorded in vital registration systems. In The Lancet HIV, Haidong Wang and colleagues6GBD 2015 HIV CollaboratorsEstimates of global, regional, and national incidence, prevalence, and mortality of HIV, 1980–2015: the Global Burden of Disease Study 2015.Lancet HIV. 2016; (published online July 19.)http://dx.doi.org/10.1016/S2352-3018(16)30087-XGoogle Scholar have refined this methodology and data sources, and present updated estimates of HIV mortality, prevalence, and incidence from 1980 to 2015 at the global, regional, and national level. The GBD 2015 Study is far more than an update; it offers an opportunity to discuss several important issues about what remains to be done to track HIV epidemics. First, health estimates are subject to substantial revision.3Brookmeyer R Measuring the HIV/AIDS epidemic: approaches and challenges.Epidemiol Rev. 2010; 32: 26-37Crossref PubMed Scopus (99) Google Scholar Wang and colleagues estimate that 38·8 million people (95% uncertainty interval [UI] 37·6–40·4 million) were living with HIV in 2015, an increase of almost 10 million compared with the GBD 2013 estimate of 29·2 million (28·1–31·7 million). The investigators also estimate that 2·5 million people (95% UI 2·2–2·7 million) were newly infected with HIV in 2015, an increase of 40% compared with the 2013 estimate of 1·8 million (1·7–2·1 million). Notably, in 2007, UNAIDS revised downwards the estimates of people living with HIV from 39·5 million to 33·2 million, and HIV incidence from 4·3 million to 2·5 million.3Brookmeyer R Measuring the HIV/AIDS epidemic: approaches and challenges.Epidemiol Rev. 2010; 32: 26-37Crossref PubMed Scopus (99) Google Scholar These revisions result from improvements in data sources and estimation methods, and do not reflect real changes in the HIV epidemics, but they also suggest that a degree of caution is warranted in interpretion of the results from these global approaches. Previously, the GDB 2013 Study concluded that the HIV epidemic was much smaller than previously thought—a difference of 6 million was found between the GBD 2013 and UNAIDS 2013 estimates of HIV prevalence, with no overlap in the uncertainty intervals.5Murray CJ Ortblad KF Guinovart C et al.Global, regional, and national incidence and mortality for HIV, tuberculosis, and malaria during 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013.Lancet. 2014; 384: 1005-1070Summary Full Text Full Text PDF PubMed Scopus (694) Google Scholar In the present update, the number of people living with HIV in 2015 was slightly, albeit not significantly, higher in GBD estimates for 2015, than in UNAIDS estimates for 2015 (36·7 million, 95% UI 34·0–39·8 million).7UNAIDSGlobal AIDS update 2016.http://www.unaids.org/sites/default/files/media_asset/global-AIDS-update-2016_en.pdfGoogle Scholar The similarity in the estimates for overall HIV prevalence in 2015 hides some important differences between them for most regions of the world. These differences need further investigation. The second issue is can we infer recent trends in incidence from the number of deaths due to the disease of interest? For HIV, the answer is no. There is a prolonged time lag between infection with HIV and death caused by HIV (on average 11 years without treatment8Wandel S Egger M Rangsin R et al.Duration from seroconversion to eligibility for antiretroviral therapy and from ART eligibility to death in adult HIV-infected patients from low and middle-income countries: collaborative analysis of prospective studies.Sex Transm Infect. 2008; 84: i31-i36PubMed Google Scholar and much longer with treatment1Wandeler G Johnson LF Egger M Trends in life expectancy of HIV-positive adults on antiretroviral therapy across the globe: comparisons with general population.Curr Opin HIV AIDS. 2016; (published online May 31.)https://doi.org/10.1097/COH.0000000000000298Crossref PubMed Scopus (214) Google Scholar). This lag means that the number of HIV deaths could, at best, provide information about trends in HIV incidence a decade ago, but not about recent trends. Furthermore, with increased access to multiple classes and lines of highly effective ART, mortality directly related to HIV will hopefully become rare, as is already the case in most high-income countries, and thus use of trends in HIV mortality to inform trends in HIV incidence will prove ineffective. We,9Supervie V Archibald CP Costagliola D et al.GBD 2013 and HIV incidence in high income countries.Lancet. 2015; 385: 1177Summary Full Text Full Text PDF PubMed Scopus (5) Google Scholar and others,4Hallett TB Zaba B Stover J et al.Embracing different approaches to estimating HIV incidence, prevalence and mortality.AIDS. 2014; 28: S523-S532Crossref PubMed Scopus (13) Google Scholar have already drawn attention to this issue and showed that the GBD estimates of HIV incidence for the regions of North America, Europe, central Asia, and Australasia were highly unrealistic. Not surprisingly, the issue remains present in Wang and colleagues' update, because the main changes introduced in the 2015 study did not include changes to the methods used for these regions. The GBD estimates of HIV incidence are significantly lower (two to ten times) than the reported number of newly diagnosed HIV cases for most countries in these regions (table), while the number of new HIV infections is expected to be close to or greater than the reported number of newly diagnosed HIV cases, since not every infection is immediately diagnosed.16Mocroft A Lundgren J Antinori A et al.Late presentation for HIV care across Europe: update from the Collaboration of Observational HIV Epidemiological Research Europe (COHERE) study, 2010 to 2013.Euro Surveill. 2015; (published online Nov 18.)https://doi.org/10.2807/1560-7917.ES.2015.20.47.30070Crossref PubMed Scopus (11) Google ScholarTableNewly diagnosed HIV cases reported to surveillance systems compared with GBD estimatesNewly diagnosed HIV cases in 2014GBD estimates of the number of new HIV infections in 2015 (95% UI)6GBD 2015 HIV CollaboratorsEstimates of global, regional, and national incidence, prevalence, and mortality of HIV, 1980–2015: the Global Burden of Disease Study 2015.Lancet HIV. 2016; (published online July 19.)http://dx.doi.org/10.1016/S2352-3018(16)30087-XGoogle ScholarRussia15Country Coordinating Mechanism RussiaThe fight against HIV/AIDS and tuberculosis: Country Coordinating Mechanism in action.http://www.hivrussia.ru/skm/info_en.shtmlGoogle Scholar85 25257 340 (32 750–102 270)USA11Centers for Disease Control and PreventionHIV surveillance report.http://www.cdc.gov/hiv/pdf/library/reports/surveillance/cdc-hiv-surveillance-report-us.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.44 07323 040 (17 680–29 960)Ukraine10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar15 79613 490 (9920–18 670)France14InVSDécouvertes de séropositivité VIH et de sida. Point épidémiologique du 1er avril.http://www.invs.sante.fr/Dossiers-thematiques/Maladies-infectieuses/VIH-sida-IST/Infection-a-VIH-et-sida/Actualites/Decouvertes-de-seropositivite-VIH-et-de-sida.-Point-epidemiologique-du-1er-avril-2016Date: 2016Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.6584960 (360–2040)UK10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.61412060 (1660–2540)Italy10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar36951960 (760–4190)Germany10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar35251760 (650–3660)Spain10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar33662350 (990–4760)Kazakhstan10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar23501630 (880–2640)Canada12Public Health Agency of CanadaHIV and AIDS in Canada: surveillance report to December 31st.http://www.phac-aspc.gc.ca/aids-sida/publication/survreport/2013/dec/index-eng.phpDate: 2013Google Scholar2090†Number of newly diagnosed HIV cases in 2013.1110 (180–2810)Turkey10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar1812720 (280–1280)Belarus10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar18111370 (760–2290)Australia13The Kirby InstituteHIV, viral hepatitis and sexually transmissible infections in Australia. Annual surveillance report 2015. Available at.http://kirby.unsw.edu.au/sites/default/files/hiv/resources/ASR2015_v4.pdfGoogle Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.1333390 (150–840)Poland10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.1061410 (130–670)Belgium10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.1039210 (60–470)Tajikistan10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.985320 (140–610)Portugal10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar9202220 (530–4910)Netherlands10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.831200 (70–470)Moldova10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar831540 (310–920)Romania10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.791450 (150–690)Greece10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.71450 (30–90)Kyrgyzstan10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.645320 (160–620)Azerbaijan10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.604360 (170–580)Georgia10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.536150 (80–250)Switzerland10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.515200 (50–450)Israel10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.477170 (50–350)Ireland10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.35960 (10–140)Sweden10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.35080 (30–150)Latvia10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar347170 (50–350)Armenia10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.33270 (30–130)Estonia10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.291110 (60–190)Hungary10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.27160 (40–80)Norway10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.26850 (20–110)Denmark10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar256130 (30–300)Bulgaria10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar247140 (60–260)Austria10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar235310 (100–710)Czech Republic10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.23240 (10–70)Finland10European Centre for Disease Prevention and ControlHIV/AIDS surveillance in Europe.http://ecdc.europa.eu/en/publications/Publications/hiv-aids-surveillance-in-Europe-2014.pdfDate: 2014Google Scholar*Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.18130 (10–80)Countries are ranked according to the reported number of newly diagnosed HIV cases. GBD=Global Burden of Disease. UI=uncertainty interval.* Countries in which the GBD uncertainty intervals do not include the value of the reported number of newly diagnosed HIV cases.† Number of newly diagnosed HIV cases in 2013. Open table in a new tab Countries are ranked according to the reported number of newly diagnosed HIV cases. GBD=Global Burden of Disease. UI=uncertainty interval. The growing effect of ART on the survival period from infection to death also challenges the other modelling approaches relying on HIV prevalence data to estimate trends in HIV incidence, including the UNAIDS approach.17Hallett TB Estimating the HIV incidence rate: recent and future developments.Curr Opin HIV AIDS. 2011; 6: 102-107Crossref PubMed Scopus (40) Google Scholar Indeed, the connection between prevalence and incidence becomes weaker with increasing use of ART and the resulting increases in life expectancy of people living with HIV. Moreover, the way in which ART interferes with prevalence and incidence trends is dependent on many factors, such as level of access to ART, heterogeneity in access, timing of initiation, adherence and response to ART, and survival and HIV transmission on ART. The scarcity of high-quality, country-specific data hampers adequate accounting for all these factors in these modelling approaches. Similar issues were encountered with the first generation of back-calculation methods, which relied on AIDS case reporting and could no longer be used after ART became available.18Working Group on Estimation of HIV Prevalence in EuropeHIV in hiding: methods and data requirements for the estimation of the number of people living with undiagnosed HIV.AIDS. 2011; 25: 1017-1023PubMed Google Scholar Therefore, in a context with high or increasing access to ART, incidence estimates should focus neither on HIV mortality data nor on prevalence data. In many high-income countries, high-quality HIV surveillance systems exist based on case reporting of all new HIV diagnoses. These data are used to estimate country incidence on the basis of two main statistical approaches: extended back-calculation models of newly diagnosed HIV cases or biomarker approaches requiring data for recent infections.9Supervie V Archibald CP Costagliola D et al.GBD 2013 and HIV incidence in high income countries.Lancet. 2015; 385: 1177Summary Full Text Full Text PDF PubMed Scopus (5) Google Scholar, 18Working Group on Estimation of HIV Prevalence in EuropeHIV in hiding: methods and data requirements for the estimation of the number of people living with undiagnosed HIV.AIDS. 2011; 25: 1017-1023PubMed Google Scholar These methods can additionally be used to provide estimates at a subnational level to better characterise the epidemic and target adequate interventions.19Marty L, Cazein F, Pillonel J et al. Mapping the HIV epidemic to improve prevention and care: the case of France. 21st International AIDS Conference; Durban, South Africa; July 18–22, 2016. TUAC0203 (abstr).Google Scholar HIV surveillance systems collecting data for ART uptake exist in many low-income and middle-income countries. Although these data are currently underused, new extensions of the back-calculation approach could be developed to derive incidence estimates. Continued production of timely and reliable data for HIV prevalence and mortality is important and, in that sense, the contributions of both UNAIDS and GBD are highly valuable. However, without timely and reliable assessment of HIV incidence, achievement of the 90-90-90 UNAIDS target and ending of the HIV epidemic will be challenging. It is time to make the right investments.20Boerma JT Stansfield SK Health statistics now: are we making the right investments?.Lancet. 2007; 369: 779-786Summary Full Text Full Text PDF PubMed Scopus (110) Google Scholar DC was a member of the French Gilead HIV board up to 2015; has given lectures and done post-marketing studies for Janssen-Cilag, Merck-Sharp & Dohme-Chibret, and ViiV; has received travel, accommodation, or meeting expenses from Gilead, ViiV, Janssen-Cilag; and is currently a consultant of Innavirvax. VS declares no competing interests. VS thanks the National Agency of Research on AIDS and Viral Hepatitis for funding support. Estimates of global, regional, and national incidence, prevalence, and mortality of HIV, 1980–2015: the Global Burden of Disease Study 2015Scale-up of ART and prevention of mother-to-child transmission has been one of the great successes of global health in the past two decades. However, in the past decade, progress in reducing new infections has been slow, development assistance for health devoted to HIV has stagnated, and resources for health in low-income countries have grown slowly. Achievement of the new ambitious goals for HIV enshrined in Sustainable Development Goal 3 and the 90-90-90 UNAIDS targets will be challenging, and will need continued efforts from governments and international agencies in the next 15 years to end AIDS by 2030. Full-Text PDF Open Access

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.