Faible masse musculaire a été peu étudiée chez les patients avant le stade terminal de la maladie rénale chronique (MRC). Nous avons évalué la masse musculaire à partir de la créatininurie des 24h pour étudier ses déterminants, son évolution avec le déclin de la fonction rénale ainsi que ses liens avec les risques de progression vers l’insuffisance rénale terminale traitée (IRTT) et de décès avant IRTT. Dans la cohorte NephroTest incluant 1429 patients avec une MRC stades 1 à 4, le débit de filtration glomérulaire a été mesuré par la clairance du 51Cr-EDTA (DFGm) et estimé par l’équation CKD EPI (DFGe). La créatininurie moyenne à l’inclusion diminuait de 15.3±3.1 à 12.1±3.3 mmol/24 chez les hommes et de 9.6±1.9 à 7.6±2.5 chez les femmes, pour une baisse du DFGm de ≥ 60 à < 15 mL/min/1.73 m2. Être plus âgé, avoir un diabète, un faible IMC ou un niveau faible de protéinurie et d’apports protidiques était associé à un niveau faible de créatininurie. Un déclin annuel du DFGm de 5 mL/min/1.73 m2 était lié à une baisse de créatininurie, indépendamment de ces déterminants. Au cours d’un suivi médian de 3.6 ans, 229 patients ont développé une IRTT, et 113 sont décédés avant IRTT. Après ajustement sur les facteurs de confusion, le hasard ratio (HR) était de 1.6 (0.88-2.9) pour le risque de décès et de 0.60 (0.39-0.91) pour le risque d’IRTT, dans le 1er vs 4ème quartile de créatininurie. La baisse de la créatininurie apparait précocement dans la MRC et est liée au décès avant dialyse. La diminution du risque d’IRTT pourrait s’expliquer par un démarrage plus tardif de la dialyse en raison d’une surestimation du DFGm par le DFGe chez les patients avec une faible créatininurie. / Mainly described in patients on dialysis, muscle wasting has received little attention in early stage chronic kidney disease (CKD). We used 24-hour creatininuria to assess determinants of low muscle mass and its putative associations with CKD outcomes, using data from the NephroTest cohort, including 1429 non-dialysis patients with CKD stages 1 to 5. Kidney function was assessed with both measured (mGFR, by 51Cr-EDTA renal clearance) and estimated glomerular filtration rate (eGFR, by CKD-EPI equation). End-stage renal disease (ESRD) and pre-ESRD death were the main studied outcomes. The mean baseline creatininuria decreased from 15.3±3.1 to 12.1±3.3 mmol/24 h in men and from 9.6±1.9 to 7.6±2.5 in women, when mGFR fell from ≥ 60 to < 15 mL/min/1.73 m2. Other determinants of low creatininuria were an older age, diabetes, a lower body mass index, a lower level of proteinuria or protein intake. A fast annual decline in mGFR of 5 mL/min/1.73 m2 was linked with a 2-fold decrease in creatininuria, independent of changes in protein intake and other determinants of muscle mass. Over a median follow-up of 3.6 years, 229 patients developed ESRD and 113 patients died before ESRD. After adjustment for confounders, patients with low muscle mass showed a significantly higher risk for pre-ESRD death (HR 1.6, 95% CI 0.88-2.9), but a lower risk for ESRD (HR 0.60, 95% CI 0.39-0.91). The latter was reversed (HR 1.5, 95% CI 1.01-2.4) when mGFR was replaced by eGFR. Decrease in 24-hour creatininuria may appear early in CKD patients, is related to pre-ESRD death. The lower risk for ESRD may reflect later dialysis start due to overestimation of true GFR by eGFR in patients with low muscle mass.

Foi estudado o ritmo de filtração glomerular (RFG) de pacientes com câncer bem diferenciado da tireoide submetidos à radioiodoterapia (RIT). O estudo avaliou o RFG durante estímulo do hormônio estimulador da tireoide (TSH) por suspensão da reposição hormonal tireoidiana (RHT) ou no uso do hormônio estimulador da tireoide recombinante humano (TSHrh), correlacionou o RFG com o perfil hormonal tireoidiano, avaliou o RFG durante e na semana após a RIT, avaliou o RFG e a dose efetiva de radiação para corpo inteiro e correlacionou métodos de estimativa de RFG. Vinte e oito pacientes incluídos em estudo clínico randomizado não cego foram divididos em dois grupos de 14 pacientes, sendo o grupo A (GA) submetido à suspensão da RHT e o grupo B (GB) ao uso do TSHrh. Os pacientes tiveram antes e após o estímulo do TSH a determinação do RFG por 51Cr-EDTA e coletas séricas do perfil hormonal tireoidiano e creatinina, albumina e ureia, e, após a RIT, colheram exames séricos de creatinina, albumina e ureia, e tiveram estimadas suas doses efetivas de corpo inteiro. Os exames de creatinina, albumina e ureia foram utilizados para estimar o RFG pelas equações de creatinina sérica, Modified Diet in Renal Disease (MDRD), e Cockcroft-Gault. O GA apresentou, pelo 51Cr-EDTA, variação de -18,5% do RFG de 94,4±18,6 mL/min antes da suspensão da RHT para 76,2±15,7 mL/min (p=0,0002) e o GB apresentou pelo 51Cr-EDTA variação de 4% do RFG de 90,8±18,4 mL/min antes do TSHrh para 92,6±15,2 mL/min (p=0,64). O RFG variou significativamente só no GA, sem apresentar proporcionalidade entre as variações do hormônio tireoidiano e do RFG. Não houve correlação do RFG com elevação do TSH. Por equações baseadas em creatinina, houve, no GA, queda do RFG durante toda a suspensão da RHT e estabilidade após o retorno da RHT, e, no GB, houve estabilidade do RFG durante todo o estudo. A dose efetiva de corpo inteiro não apresentou diferenças significativas entre os grupos (p=0,76). Na comparação entre o 51Cr-EDTA e as equações para estimativa de RFG, a correlação de Pearson foi de 0,78 para creatinina sérica, 0,79 para MDRD e 0,66 para CockcroftGault, e a comparação das variações do RFG observadas no GA entre o 51Cr-EDTA e a equação por creatinina sérica foram estatisticamente diferentes. Concluiu-se que o RFG apresenta redução na suspensão da RHT, relacionado ao hipotireoidismo e não à elevação de TSH, voltando a estabilizar após retorno da RHT, e que não varia no uso do TSHrh, que a dose efetiva de corpo inteiro não varia entre os grupos proporcionalmente ao RFG, e que a melhor correlação foi do 51Cr-EDTA com a equação MDRD / Glomerular filtration rate (GFR) was studied in well differentiated thyroid cancer patients referred for radioiodine therapy (RIT). The study evaluated GFR during thyroid stimulating hormone (TSH) stimulation after thyroid hormone withdrawal (THW) or after recombinant human thyroid stimulating hormone (rhTSH), correlated GFR with thyroid hormone profile, evaluated GFR during and in the week after RIT, evaluated GFR and whole body radiation effective dose, and correlated different methods for GFR determination. 28 patients were included in a non-blinded randomized clinical trial and divided in two groups of 14 patients, being group A (GA) stimulated by THW and group B (GB) stimulated by rhTSH. Patients had GFR determined by 51Cr-EDTA, as well as serum thyroid hormone profile, creatinine, albumin and urea before and after TSH stimulation, and after RIT had determined their serum creatinine, albumin and urea and whole body radiation effective dose. Creatinine, albumin and urea were used to estimate GFR by serum creatinine, Modified Diet in Renal Disease (MDRD), and Cockcroft-Gault equations. GA presented a -18,5% GFR variation by 51CrEDTA varying from 94,4 ± 18,6 mL/min before THW to 76,2±15,7 mL/min after THW (p=0,0002) while GB presented a 4% GFR variation by 51Cr-EDTA varying from 90,8 ± 18,4 mL/min before TSHrh to 92,6 ± 15,2 mL/min after rhTSH (p=0,64). GFR significantly varied only in GA without presenting proportionality with thyroid hormone variation. There was no correlation between rise in TSH levels and GFR. Creatinine equations demonstrated a sustained reduction in GFR during THW and GFR stability after thyroid hormone reposition, while GB presented stable GFR during the whole study. Whole body radiation effective dose didn\'t present significant differences between the two groups (p=0,76). Comparing 51Cr-EDTA and GFR estimative equations presented Pearson correlation score of 0,78 for serum creatinine, 0,79 for MDRD and 0,66 for Cockcroft-Gault, while comparison between variances in GA between 51Cr-EDTA e serum creatinine equation was significantly different. In conclusion GFR presents a reduction during THW related to hypothyroidism and not to TSH rise and stabilizing after thyroid hormone therapy, GFR does not vary during rhTSH, whole body radiation effective dose does not vary between the two groups proportionally to GFR, and that MDRD equation had the best correlation with 51Cr-EDTA

Pacientes com diagnósticos oncológicos diversos sob terapia com agentes nefrotóxicos, foram avaliados através de procedimentos radioisotópicos, em três momentos diferentes. Os achados radioisotópicos foram comparados com as avaliações laboratoriais rotineiras. Não mostraram alterações estatisticamente significativas os seguintes parâmetros: uréia, creatinina, sedimentos anormais e índice de Crockoft-Gault. Concluiu-se que a avaliação da função renal com os métodos rotineiramente utilizados na prática oncológica não foram estatisticamente significativas. Dentre os estudos utilizando radionuclídeos, o DTPA-99mTc e DMSA-99mTc, evidenciaram alterações, enquanto o EC-99mTc, não as detectou / Patients with different oncologic diagnoses under treatment with nephrotoxic drugs were evaluated by radioisotopic agents at three different moments. Radioisotopic data were compared with biochemical routine tests. Concerning laboratorial parameters like serum creatinine, urea and Cockroft Gault index, no statistically significant changes were observed. Possible to conclude that renal function evaluation with methods routinely used in oncological practice did not reveal any statistically significant change. Radioisotopic studies using 99mTc-DTPA and 99mTc-DMSA showed considerable alterations, however with 99mTc-EC no significant changes were evidenced

Pacientes com diagnósticos oncológicos diversos sob terapia com agentes nefrotóxicos, foram avaliados através de procedimentos radioisotópicos, em três momentos diferentes. Os achados radioisotópicos foram comparados com as avaliações laboratoriais rotineiras. Não mostraram alterações estatisticamente significativas os seguintes parâmetros: uréia, creatinina, sedimentos anormais e índice de Crockoft-Gault. Concluiu-se que a avaliação da função renal com os métodos rotineiramente utilizados na prática oncológica não foram estatisticamente significativas. Dentre os estudos utilizando radionuclídeos, o DTPA-99mTc e DMSA-99mTc, evidenciaram alterações, enquanto o EC-99mTc, não as detectou / Patients with different oncologic diagnoses under treatment with nephrotoxic drugs were evaluated by radioisotopic agents at three different moments. Radioisotopic data were compared with biochemical routine tests. Concerning laboratorial parameters like serum creatinine, urea and Cockroft Gault index, no statistically significant changes were observed. Possible to conclude that renal function evaluation with methods routinely used in oncological practice did not reveal any statistically significant change. Radioisotopic studies using 99mTc-DTPA and 99mTc-DMSA showed considerable alterations, however with 99mTc-EC no significant changes were evidenced

The aim of this dissertation thesis was to study the factors affecting drug distribution and elimination and to use these factors to individualize dosing. The work consists of three thematic areas: estimation of the volume of distribution and subsequent dosing of selected drugs (vancomycin, amikacin, phenobarbital) using body size descriptors; estimation of clearance and subsequent dosing of selected drugs (vancomycin, amikacin, phenobarbital, perindopril) using renal function status markers; and the impact of drug interactions on the distribution and elimination of phenobarbital. The thesis summarizes original papers on these topics. Individual pharmacokinetic parameters were calculated for each patient based on their demographic and clinical characteristics, dosing records and measured serum drug levels. The relationships between distribution volume/drug clearance and body size descriptors/renal functional status markers were examined by regression analysis. Vancomycin volume of distribution was best predicted by the total body weight. Loading dose of 10.7 mg/kg of total body weight was optimal in patients taking continuous vancomycin and would lead to reducing of median time to reach target concentrations from 17 to 1 hour. On the contrary, amikacin volume of distribution was most associated...

Vários estudos mostram que na doença renal policística autossômica dominante os cistos surgem a partir de um mecanismo de "dois-golpes". A patogênese das manifestações não-císticas, contudo, é pouco compreendida. Neste estudo usamos uma linhagem de camundongos endogâmica com uma mutação nula em Pkd1, onde animais heterozigotos apresentam formação cística renal mínima até 40 semanas de idade. O clearance de inulina e o número de glomérulos foram menores em machos Pkd1+/- que Pkd1+/+, enquanto o volume glomerular médio foi maior em heterozigotos. A excreção urinária de NO2/NO3 não diferiu significantemente entre os dois grupos. Avaliamos a osmolalidade urinária máxima em machos e fêmeas Pkd1+/- and Pkd1+/+, porém não foi detectada diferença significante entre os grupos heterozigoto e selvagem. Nossos resultados oferecem evidência direta de que a haploinsuficiência de Pkd1 resulta em anormalidades anatômicas e funcionais renais e sugerem que o estado haploinsuficiente de Pkd1 possa resultar na redução do número de néfrons por diminuir a ramificação tubular renal durante a nefrogênese / Several studies show that in autosomal dominant polycystic kidney disease cysts arise through a "two-hit" mechanism. The pathogenesis of non-cystic features, however, is poorly understood. In this study we used an inbred mouse line with a null mutation of Pkd1, where heterozygotes had minimal renal cyst formation up to 40 weeks of age. Inulin clearance and the number of glomeruli were lower in Pkd1+/- than in Pkd1+/+ males, while a higher average glomerular volume was observed in heterozygotes. The urinary excretion of NO2/NO3 did not significantly differ between the two groups. Maximal urinary osmolality was evaluated in Pkd1+/- and Pkd1+/+ males and females, but no significant difference was detected between the heterozygous and the wild type groups. Our results provide direct evidence that haploinsufficiency for Pkd1 results in anatomic and functional abnormalities of the kidney and suggest that Pkd1 haploinsufficiency may result in a reduced number of nephrons by diminishing renal tubule branching during nephrogenesis

INTRODUÇÃO: Em pacientes transplantados renais, a paratireoidectomia está associada à piora aguda da função renal. Os efeitos agudos e crônicos da paratireoidectomia sobre a filtração glomerular foram pouco estudados em Hiperparatireoidismo Primário (HPTP). MÉTODO E CASUÍSTICA: Neste estudo retrospectivo de coorte, foram estudados 494 pacientes submetidos a paratireoidectomia por HPTP, entre os anos de 2007 e 2016. Variações agudas da creatinina foram aferidas diariamente na internação, até o 4o pós-operatório, sendo classificados conforme os critérios de KDIGO para IRA. Dados bioquímicos incluíram dosagem sérica de creatinina, cálcio iônico e total, paratormônio (PTH) e 25-OH vitamina D. A taxa de filtração glomerular foi estimada a partir da equação CKD-EPI. Foram comparados dados de função renal pré e pós-operatórios até 5 anos de seguimento. RESULTADOS: Dos 494 pacientes, 391 (79,1%) eram mulheres e 422 (85,4%) de cor branca. A causa mais comum de HPTP foi adenoma de paratireóide (351, 71,1%) e a mediana de idade foi de 58 anos. As medianas (Q1-Q3) de creatinina, PTH e cálcio total séricos foram de: 0,81 mg/dL (0,68-1,01), 154,5 pg/mL (106-238,5) e 10,9 mg/dL (10,3-11,5) respectivamente. A mediana de eGFR préoperatória foi de 86 mL/min x 1,73m2. No período agudo, houve redução mediana de 26 mL/min x 1,73m2 na eGFR (p < 0,0001), que representou -27,44% (±19,12%) de variação aguda da eGFR. De acordo com os critérios de IRA, 41,1% dos pacientes tiveram IRA estágio 1, 5,9% estágio 2 e 1,8% estágio 3. Outros 223 pacientes (45,1%) tiveram elevação da creatinina porém não preencheram critérios de IRA. Na análise univariada foram observadas correlações fracas, porém significativas, entre o percentual de variação aguda de eGFR e os seguintes fatores pré-operatórios: idade, PTH, cálcio e creatinina. Uma redução definitiva da eGFR foi observada em 60,7% dos pacientes, após 12 meses de seguimento. CONCLUSÃO: Houve significativa redução aguda da função renal após paratireoidectomia por HPTP, sendo que quase metade dos pacientes preencheram critérios de IRA. Observou-se importante recuperação da eGFR no primeiro mês de pós-operatório, podendo ocorrer algum grau de perda definitva de função renal / INTRODUCTION: In kidney transplant patients, parathyroidectomy is associated with acute decrease in renal function. Acute and chronic effects of parathyroidectomy on renal function have not been as extensively studied in primary hyperparathyroidism (PHPT). PATIENTS AND METHODS: Retrospective cohort study of 494 patients undergoing parathyroidectomy for PHPT. Acute renal changes were evaluated daily until day 4 post parathyroidectomy, and stratified according to acute kidney injury (AKI) criteria. Biochemical assessment included serum creatinine, total and ionized calcium, PTH, and 25-hydroxyvitamin D (25OHD). EGFR were calculated using the CKD-EPI equation. We compared preoperative and postoperative renal function up to 5 years of follow-up. RESULTS: 391 (79.1%) patients were female and 422 (85.4%) were non-African American. Median age was 58 years old. Median (interquartile range) preoperative serum creatinine, PTH and total calcium were 0.81 mg/dL (0.68- 1.01), 154.5 pg/mL (106-238.5), and 10.9 mg/dL (10.3-11.5) respectively. Median (interquartile range) preoperative eGFR was 86 mL/min/1.73m2 (65-101.3). After surgery the median acute decrease in eGFR was 26 mL/min/1.73m2 (p < 0.0001). Acutely, 41.1% patients developed AKI stage 1, 5.9% AKI stage 2 and 1.8% AKI stage 3. Acute eGFR decrease (%) correlated with age, PTH, calcium and preoperative creatinine, in univariate analysis. Permanent reduction in eGFR occurred in 60.7 % of the patients, after acute episode. CONCLUSION: There is a significant acute impairment in renal function after parathyroidectomy for PHPT and almost half of patients meet the criteria for AKI. Significant eGFR recovery was observed during first month after surgery, but a small permanent reduction may occur

Cette thèse avait pour but d'illustrer l'intérêt de méthodes statistiques avancées lorsqu'on s'in­ téresse aux associations entre différents facteurs et la progression de la maladie rénale chronique (MRC). Dans un premier temps, une revue de la littérature a été effectuée alin d'identifier les méthodes classiquement utilisées pour étudier les facteurs de progression de la MRC ; leurs limites et des méthodes permettant de mieux prendre en compte ces limites ont été discutées. Notre second travail s'est concentré sur les analyses de données de survie et la prise en compte de la censure par intervalle, qui survient lorsque l'évènement d'intérêt est la progression vers un stade spécifique de la MRC, et le risque compétitif avec le décès. Une comparaison entre des modèles de survie standards et le modêle illness-death pour données censurées par intervalle nous a permis d'illustrer l'impact de la modélisation choisie sur les estimations à la fois des effets des facteurs de risque et des probabilités d'évènements, à partir des données de la cohorte NephroTest. Les autres travaux ont porté sur les analyses de données longitudinales de la fonction rénale. Nous avons illustré l'intérêt du modèle linéaire mixte dans ce contexte et présenté son extension pour la prise en compte de sous-populations de trajectoires de la fonction rénale différentes. Nous avons ainsi identifier cinq classes, dont une avec un déclin très rapide et une autre avec une amélioration de la fonction rénale au cours du temps. Des perspectives de travaux liés à la prédiction permettent enfin de lier les deux types d'analyses présentées dans la thèse. / The objective of this thesis was to illustrate the benefit of using advanced statistical methods to study associations between risk factors and chrouic kidney disease (CKD) progression. In a first time, we conducted a literature review of statistical methods used to investigate risk factors of CKD progression, identified important methodological issues, and discussed solutions. In our sec­ ond work, we focused on survival analyses and issues with interval-censoring, which occurs when the event of interest is the progression to a specifie CKD stage, and competing risk with death. A comparison between standard survival models and the illness-death mode! for interval-censored data allowed us to illustrate the impact of modeling on the estimates of both the effects of risk factors and the probabilities of events, using data from the NephroTest cohort. Other works fo­ cused on analysis of longitudinal data on renal function. We illustrated the interest of linear mixed mode! in this context and presented its extension to account for sub-populations with different trajectories of renal function. We identified five classes, including one with a strong decline and one with an improvement of renal function over time. Severa! perspectives on predictions bind the two types of analyses presented in this thesis.

Diabetic vasculopathy (DV) is the most important consequence of chronic hyperglycemia in patients with diabetes mellitus (DM). This thesis explores the interaction of blood, bone and kidney in the pathogenesis of DV by i) reviewing the current understanding of pathogenesis of macrovascular and microvascular diseases in DM to identify gaps in literature and generate hypotheses relating to various facets of DV ii) undertaking a series of prospective studies to examine these hypotheses iii) analysing the findings and integrating any new information obtained from the clinical studies into the current knowledge base and iv) generating hypotheses upon which future work might be based. The literature search was carried out with the aim of understanding current concepts of pathogenesis of DV and its potential modulators. The original reviews resulting from this process are presented in chapters 2 to 4. A series of pilot studies reported in chapters 7 to 11, were then carried out to interrogate hypotheses originating from this process. The first study was carried out in healthy individuals to define the biological variation of potential modulators of DV, namely erythropoietin (EPO), parathyroid hormone, 25 hydroxyvitamin D and 1, 25-dihydroxyvitamin D to facilitate the design and interpretation of subsequent studies. It revealed a wide biological variation of these modulators in the healthy population thus,emphasizing the need to have a control group in the subsequent study population. To examine whether tubulointerstitial dysfunction occurs before the onset of microalbuminuria, a measurement of the above mentioned parameters was carried out along with markers of tubulointerstitial injury in patients with type 1 and type 2 DM without microalbuminuria and in non-diabetic controls. It was found that tubulointerstitial dysfunction with low levels of EPO and 1, 25-dihydroxyvitamin D and higher excretion of tubular injury markers, occurs before the onset of microalbuminuria. Subsequently, diabetic and nondiabetic chronic kidney disease (CKD) patients with EPO deficiency anaemia were examined to study the effects of EPO therapy on the excretion of tubular injury markers. However, in these patient groups, we were unable to demonstrate an effect of EPO therapy on the markers of tubular injury in spite of a beneficial haematological response. To examine whether vascular calcification (VC) and bone mineral density (BMD) were linked in patients with diabetes mellitus and to explore their relationship to modulators of DV, an assessment of VC and BMD was undertaken in patients with type 2 DM with different degrees of proteinuria and normoalbuminuria. VC was assessed by CT scan and BMD by a DEXA scan. Modulators of DV were measured including serum Osteoprotegerin (OPG) and receptor activator of nuclear factor kappa-b-ligand (RANKL). The findings were i) a high prevalence of VC and osteopenia in normoalbuminuric type 2 DM patients with normal serum creatinine ii) a weak inverse relationship between VC and osteopenia iii) proteinuric patients had worse VC but not osteopenia iv) weak relationships between OPG levels and both VC and osteopenia, masked by age in multivariate analysis. The final study examined the relationship between modulators of DV, including OPG and RANKL, and the degree of CKD. It was found that abnormalities of OPG and RANKL occur before the onset of microalbuminuria and progress with deterioration of renal function. Compared to nondiabetics, DM patients have higher OPG levels in the predialysis phase and lower levels in haemodialysis phase, a phenomenon that might indicate endothelial exhaustion in dialysis patients with DM. The derangements associated with DV seem to occur earlier than previously thought. Further work is required to untangle these complexities and to define the contribution of factors such as the adverse blood milieu, the vasculature, abnormal bone and mineral metabolism, and early tubulointerstitial damage. The findings from the studies reported here may help in the formulation of new hypotheses, which might contribute to future work in this area.

Le virus de l'immunodéficience humaine (VIH) est à l’origine d’une infection chronique, elle-même responsable du développement du syndrome d'immunodéficience acquise (SIDA), un état de grande vulnérabilité où le corps humain est à la merci d’infections opportunistes pouvant s’avérer fatales. Aujourd’hui, 30 ans après la découverte du virus, même si aucun vaccin n’a réussi à contrôler la pandémie, la situation s’est grandement améliorée. Conséquemment à l’arrivée de traitements antirétroviraux hautement actifs (HAART) à la fin des années 1990, la mortalité associée au VIH/SIDA a diminué et un plus grand nombre de personnes vivent maintenant avec l'infection. La présente thèse avait pour objectif d’aborder trois situations problématiques, en dépit de l’efficacité reconnue des HAART, plus particulièrement la faible charge virale persistante (LLV) et sa relation avec l’échec virologique, ainsi que les effets de certains antirétroviraux (ARV) sur les fonctions rénale et hépatique. Les objectifs précis étaient donc les suivants : 1) étudier le risque d’échec virologique à long terme chez les patients sous HAART dont la charge virale est indétectable comparativement aux patients affichant une LLV persistante; 2) évaluer sur le long terme la perte de fonction rénale associée à la prise de ténofovir (TDF) 3) étudier sur le long terme l'hyperbilirubinémie associée à la prise d’atazanavir (ATV) et ses autres déterminants possibles. Afin d’atteindre les trois objectifs susmentionnés, une cohorte de 2 416 patients atteints du VIH/SIDA, suivis depuis juillet 1977 et résidant à Montréal, a été utilisée. Pour le premier objectif, les résultats obtenus ont montré un risque accru d’échec virologique établi à >1000 copies/ml d’ARN VIH chez tous les patients qui présentaient une LLV persistante de différentes catégories durant aussi peu que 6 mois. En effet, on a observé qu’une LLV de 50-199 copies/ml persistant pendant six mois doublait le risque d’échec virologique (Hazard ratio (HR)=2,22, Intervalle de confiance (CI) 95 %:1,60–3,09). Ces résultats pourraient modifier la façon dont on aborde actuellement la gestion des patients affichant une LLV, et plus particulièrement une LLV de 50-199 copies/ml, pour laquelle aucune recommandation clinique n’a encore été formulée en raison du manque de données. Pour le deuxième objectif, on a observé une augmentation du risque de perte de fonction rénale de l’ordre de 63 % (HR=1,63; 95% CI:1,26–2,10) chez les patients sous TDF comparativement aux patients traités avec d’autres ARV. La perte de fonction rénale directement attribuable à la prise de TDF, indique que cette perte est survenue au cours des premières années de l’exposition du patient au médicament. D’une perspective à long terme, cette perte est considérée comme modérée. Enfin, pour ce qui est du troisième objectif, on a constaté que l’incidence cumulative d’hyperbilirubinémie était très élevée chez les patients sous ATV, mais que cette dernière pouvait régresser lorsque l’on mettait fin au traitement. L’hyperbilirubinémie à long terme observée avec la prise d’ATV n’a été associée à aucun effet néfaste pour la santé. Dans l’ensemble, la présente thèse a permis de mieux comprendre les trois situations problématiques susmentionnées, qui font actuellement l’objet de débats au sein de la communauté scientifique, et d’éclairer sous un jour nouveau la gestion des patients séropositifs sous traitement médicamenteux. / Human immonudeficiency virus (HIV) is a virus causing a chronic infection responsible for Acquired Immunodeficiency Syndrome (AIDS), a state of vulnerability of the body where different opportunistic infections will ultimately be fatal. About 30 years after the discovery of the virus, even if no vaccine is available to control the pandemia, situation has changed for the best. With the arrival of highly active anti-retroviral therapy (HAART) in the late 90's, a reduction in HIV/AIDS mortality rate and growing number of persons living with the infection were observed. The overall objective of this thesis was to address three problematic situations, despite recognised HAART efficacy, especially low-level viremia (LLV) and its relationship with virologic failure, and the impacts of certain antiretrovirals (ARV) on kidney and hepatic functions. The specific objectives were: 1) to study the risk of virologic failure in long-term perspective in undetectable patients under HAART in comparison to patients with persistent LLV; 2) to evaluate the long-term loss of kidney function related to tenofovir (TDF) exposure 3) to evaluate long-term hyperbilirubinemia related to atazanavir (ATV) exposure and other possible determinants. In order to address the three specific objectives, a cohort of patients 2416 living with HIV/AIDS followed in Montreal since July 1977 was used. For the first objective, analyses and results shown an increased risk of virological failure defined as >1000 copies/mL of HIV RNA, for all categories of persistent LLV as soon as 6 months of persistent duration. Persistent LLV of 50-199 copies/mL for 6 months doubled the risk of virologic failure (Hazard ratio (HR)=2,22, Confidence interval (CI) 95%: 1,60-3,09). The results shed new light for the management of patients with LLV, especially for LLV of 50-199 copies/mL, for which no clinical recommendation is currently available due to a lack of data. For the second objective, an increased risk of loss of kidney function of 63% (HR=1.63; 95% CI:1.26–2.10) associated to TDF exposure in comparison to patients taking other ARV was observed. The cumulative eGFR loss directly attribuable to TDF also shown that this loss occured during the first years of exposure. This loss was mild in a long-term perspective. For the third objective, it has been shown that the cumulative incidence of hyperbilirubinemia in ATV users was very high and that regression was possible if ATV exposure was ended. Long-term hyperbilirubinemia related to ATV use was not associated with adverse health outcome. Overall, this thesis allowed a better understanding of these three problematics currently debated in scientific literature and shed new lights on management of HIV positive patients under therapy.