A New Approach for Cognitive Impairment Pattern in Chronic Kidney Disease.

BACKGROUND AND HYPOTHESIS: Chronic kidney disease (CKD) is associated with an elevated risk of neurocognitive disorders (NCDs). It remains unclear whether CKD-related NCDs have specific cognitive pattern or are earlier-onset phenotypes of the main NCDs (vascular NCDs and Alzheimer's disease). METHODS: We used the Mini Mental State Examination score (MMSE) to assess cognitive pattern in 3003 CKD patients (stage 3 to 4) followed up over 5 years in the Chronic Kidney Disease-Renal Epidemiology and Information Network (CKD-REIN) cohort. After normalizing MMSE scores to a 0-to-100 scale, the associations between the baseline estimated glomerular filtration rate (eGFR, using the CKD-EPI-creatinine formula) and changes in each MMSE domain score were assessed in linear mixed models. RESULTS: Patients (age: 67±13 years old; males: 65%, mean eGFR: 33±12 ml/min/1.73 m²) had a good baseline cognitive functions: the mean MMSE score was 26.9/30 ±2.9. After adjustment for age, sex, educational level, depression (past or present), cardiovascular risk factors, cerebrovascular disease, a lower baseline eGFR (per 10 ml/min/1.73 m²) was associated with a 0.53-point decrement (p<0.001; 95%CI [-0.98,-0.08]) for orientation, a 1.04-point decrement (p=0.03; 95%CI [-1.96,-0.13]) for attention and calculation, a 0.78-point decrement (p=0.003; 95%CI [-1.30,-0.27]) for language, and a 0.94-point decrement (p=0.02; 95%CI [-1.75,-0.13]) for praxis. Baseline eGFR was not, however, associated with significant changes over time in MMSE domain scores. CONCLUSION: A lower eGFR in CKD patients was associated with early impairments in certain cognitive domains: praxis, language and attention domains before an obvious cognitive decline. Early detection of NCD in CKD patients must be perform before clinically cognitive decline using preferably tests assessing executive, attentional functions and language than memory test. This could lead to a better management of cognitive impairment and their consequences on CKD management.

Hypothermic Perfusion Modifies the Association Between Anti-LG3 Antibodies and Delayed Graft Function in Kidney Recipients.

We previously reported associations between autoantibodies to the LG3 fragment of perlecan, anti-LG3, and a higher risk of delayed graft function (DGF) in kidney transplant recipients. Here, we aimed to determine whether some factors that modulate ischemia-reperfusion injury (IRI) can modify this association. We performed a retrospective cohort study in kidney transplant recipients in 2 university-affiliated centers. In 687 patients, we show that high pre-transplant anti-LG3 are associated with DGF when the kidney is transported on ice (odds ratio (OR): 1.75, 95% confidence interval 1.02-3.00), but not when placed on hypothermic perfusion pump (OR: 0.78, 95% CI 0.43-1.37). In patients with DGF, high pre-transplant anti-LG3 are associated with a higher risk of graft failure (subdistribution hazard ratio (SHR): 4.07, 95% CI: 1.80, 9.22), while this was not the case in patients with immediate graft function (SHR: 0.50, 95% CI 0.19, 1.29). High anti-LG3 levels are associated with a higher risk of DGF in kidneys exposed to cold storage, but not when hypothermic pump perfusion is used. High anti-LG3 are also associated with a higher risk of graft failure in patients who experience DGF, a clinical manifestation of severe IRI.

Cognitive performance is associated with glomerular filtration rate in patients with chronic kidney disease: results from the CKD-REIN cohort.

BACKGROUND: Chronic kidney disease (CKD) is associated with cognitive impairment in general population. We assessed the association between kidney and cognitive functions in patients with CKD and the influence of cardiovascular (CV) risk factors, and depression on this association. METHODS: The CKD-Renal Epidemiology and Information Network cohort included 3033 patients with CKD stages 3-4, followed for 5 years. Cognitive function was assessed with the Mini-Mental State Examination (MMSE) and estimated glomerular filtration rate (eGFR) with the CKD-Epidemiology Collaboration equation-creatinin formula. Evolution of the MMSE score over time and its association with baseline eGFR were investigated with linear mixed models. We assessed the risk of incident cognitive outcome (hospitalisation or death with relevant International Classification of Disease-10 codes), with a Cox proportional hazard model. RESULTS: The mean age was 66.8, the mean eGFR was 33 mL/min/1.73 m2 and 387 patients (13.0%) had an MMSE score below 24 at baseline. A 10 mL/min/1.73 m2 decrement of baseline eGFR was associated with a mean MMSE decrease of 0.12 (95% CI 0.04 to 0.19) after adjustment for demographic characteristics, depression, CV risk factors and disease; but baseline eGFR was not associated with MMSE temporal evolution. HR for cognitive outcome during follow-up (median 2.01 years) associated with a 10 mL/min/1.73 m2 decrement of baseline eGFR was 1.35 (1.07, 1.70) (p=0.01) after adjustment. CONCLUSIONS: In patients with CKD, lower eGFR was associated with worse cognitive performance and incident cognitive events, independently of demographics, CV risk factors and depression. TRIAL REGISTRATION NUMBER: NCT03381950.

Impact of kidney transplantation in obese candidates: a time-dependent propensity score matching study.

BACKGROUND: Although kidney transplantation (KT) is considered the best treatment for end-stage renal disease (ESRD), there are concerns about its benefit in the obese population because of the increased incidence of post-transplant adverse events. We compared patients who underwent KT versus patients awaiting KT on dialysis. METHODS: We estimated the life expectancy [restricted mean survival time (RMST)] for a 10-year follow-up by matching on time-dependent propensity scores. The primary outcome was time to death. RESULTS: In patients with a body mass index (BMI) ≥30 kg/m2 (n = 2155 patients per arm), the RMST was 8.23 years [95% confidence interval (CI) 8.05-8.40] in the KT group versus 8.00 years (95% CI 7.82-8.18) in the awaiting KT group, a difference of 2.71 months (95% CI -0.19-5.63). In patients with a BMI ≥35 kg/m2 (n = 212 patients per arm), we reported no significant difference [8.56 years (95% CI 7.96-9.08) versus 8.66 (95% CI 8.10-9.17)]. Hence we deduced that KT in patients with a BMI between 30 and 35 kg/m2 was beneficial in terms of life expectancy. CONCLUSION: Regarding the organ shortage, KT may be questionable for those with a BMI ≥35 kg/m2. These results do not mean that a BMI ≥35 kg/m2 should be a barrier to KT, but it should be accounted for in allocation systems to better assign grafts and maximize the overall life expectancy of ESRD patients.

Longitudinal Bone Loss Occurs at the Radius in CKD.

INTRODUCTION: Chronic kidney disease (CKD) exposes to an increased incidence of fragility fractures. International guidelines recommend performing bone mineral density (BMD) if the results will impact treatment decisions. It remains unknown where bone loss occurs and what would preclude the longitudinal loss in patients with CKD. Here, we aimed to investigate factors influencing BMD and to analyze the longitudinal BMD changes. METHODS: In the NephroTest cohort, we measured BMD at the femoral neck, total hip, lumbar spine, and proximal radius, together with circulating biomarkers and standardized measured glomerular filtration rate (mGFR) by 51Cr-EDTA in a subset of patients with CKD stage 1 to 5 followed during 4.3 ± 2.0 years. A linear mixed model explored the longitudinal bone loss and the relationship of associated factors with BMD changes. A total of 858 patients (mean age 58.9 ± 15.2 years) had at least 1 and 477 had at least 2 BMD measures. RESULTS: At baseline, cross-sectional analysis showed a significantly lower BMD at femoral neck and total hip and a significant higher serum parathyroid hormone (PTH) along with CKD stages. Baseline age, gender, tobacco, low body mass index (BMI), and high PTH levels were significantly associated with low BMD. Longitudinal analysis during the mean 4.3 years revealed a significant bone loss at the radius only. BMD changes at the femoral neck were associated with BMI, but not CKD stages or basal PTH levels. CONCLUSIONS: CKD is associated with low BMD and high PTH in the cross-sectional analysis. Longitudinal bone loss occurred at the proximal radius after 4.3 years.

Urgent-start dialysis in patients referred early to a nephrologist-the CKD-REIN prospective cohort study.

BACKGROUND: The lack of a well-designed prospective study of the determinants of urgent dialysis start led us to investigate its individual- and provider-related factors in patients seeing nephrologists. METHODS: The Chronic Kidney Disease Renal Epidemiology and Information Network (CKD-REIN) is a prospective cohort study that included 3033 patients with CKD [mean age 67 years, 65% men, mean estimated glomerular filtration rate (eGFR) 32 mL/min/1.73 m2] from 40 nationally representative nephrology clinics from 2013 to 2016 who were followed annually through 2020. Urgent-start dialysis was defined as that 'initiated imminently or <48 hours after presentation to correct life-threatening manifestations' according to the Kidney Disease: Improving Global Outcomes 2018 definition. RESULTS: Over a 4-year (interquartile range 3.0-4.8) median follow-up, 541 patients initiated dialysis with a known start status and 86 (16%) were identified with urgent starts. The 5-year risks for the competing events of urgent and non-urgent dialysis start, pre-emptive transplantation and death were 4, 17, 3 and 15%, respectively. Fluid overload, electrolytic disorders, acute kidney injury and post-surgery kidney function worsening were the reasons most frequently reported for urgent-start dialysis. Adjusted odds ratios for urgent start were significantly higher in patients living alone {2.14 [95% confidence interval (CI) 1.08-4.25] or with low health literacy [2.22 (95% CI 1.28-3.84)], heart failure [2.60 (95% CI 1.47-4.57)] or hyperpolypharmacy [taking >10 drugs; 2.14 (95% CI 1.17-3.90)], but not with age or lower eGFR at initiation. They were lower in patients with planned dialysis modality [0.46 (95% CI 0.19-1.10)] and more nephrologist visits in the 12 months before dialysis [0.81 (95% CI 0.70-0.94)] for each visit. CONCLUSIONS: This study highlights several patient- and provider-level factors that are important to address to reduce the burden of urgent-start dialysis.

Clinical Trial Emulation by Matching Time-dependent Propensity Scores: The Example of Estimating Impact of Kidney Transplantation.

BACKGROUND: No study to our knowledge has examined the use of observational data to emulate a clinical trial whereby patients at the time of kidney transplant proposal are randomly assigned to an awaiting transplantation or transplantation group. The main methodologic issue is definition of the baseline for dialyzed patients assigned to awaiting transplantation, resulting in the inability to use common propensity score-based approaches. We aimed to use time-dependent propensity score to better appraise the benefit of transplantation. METHODS: We studied 23,231 patients included in the French registry and on a transplant waiting list from 2005 to 2016. The main outcome was time to death. By matching on time-dependent propensity score, we obtained 10,646 pairs of recipients (transplantation group) versus comparable patients remaining on dialysis (awaiting transplantation group). RESULTS: The baseline exposure, that is, pseudo-randomization, was matching time. Median follow-up time was 3.5 years. At 10 years' follow-up, the restricted mean survival time was 8.8 years [95% confidence interval (CI) = 8.7, 8.9] in the transplantation group versus 8.2 years (95% CI = 8.1, 8.3) in the awaiting transplantation group. The corresponding life expectancy gain was 6.8 months (95% CI = 5.5, 8.2), and this corresponded to one prevented death at 10 years for 13 transplantations. CONCLUSIONS: Our study has estimated the life expectancy gain due to kidney transplantation. It confirms transplantation as the best treatment for end-stage renal disease. Furthermore, we demonstrated that this simple method should also be considered for estimating marginal effects of time-dependent exposures.

Induction therapy in kidney transplant recipients: Description of the practices according to the calendar period from the French multicentric DIVAT cohort.

BACKGROUND: There is extensive literature with comparisons between Anti-Thymocyte Globulin (ATG) and Basiliximab (BSX) as induction therapy in kidney transplant recipients. The purpose of our benchmarking study was to describe the consequences in terms of practices in 6 transplantation centers of a French prospective cohort. METHODS: We included adult patients who received a first or second kidney graft between 2013 and 2019 (n = 4157). We used logistic regressions to identify characteristics associated with the use of ATG or BSX. RESULTS: Use of ATG between the centers ranged from 41% to 75%. We observed different factors associated with the treatment decision. Compared to a first transplant, performing a second graft was the only factor significantly associated with the choice of ATG in all centers. The AUC ranged from 0.67 to 0.91, indicating that the centers seemed to define their own rules. As a result, for patients with the same low immunological risk, the probability of receiving ATG varied from 7% to 36%. We stratified the analyses according to two periods, from 2013 to 2015 and from 2016 to 2019. A similar heterogeneity was observed, and in some cases ATG indications between the centers were inverted. CONCLUSIONS: The heterogeneity of induction therapy practices did not decrease in France, even if the reated literature is prolific. This illustrates the necessity to improve the literature by using meta-analyses of recent studies stratified by graft and patient profiles.

Association between day of the week and medication adherence among adolescent and young adult kidney transplant recipients.

Disruption of usual routines may hinder adherence, increasing the risk of rejection. We aimed to compare weekend versus weekday medication adherence among adolescent and young adult kidney transplant recipients, hypothesizing poorer adherence on weekends. We examined data from the Teen Adherence in Kidney transplant Effectiveness of Intervention Trial (TAKE-IT). We assessed the 3-month run-in period (no intervention) and the 12-month intervention interval, considering a potential interaction between weekend/weekday and treatment group. Adherence was monitored using electronic pillboxes in participants 11-24 years followed in eight transplant centers in Canada and the United States. We used logistic regression with generalized estimating equations to estimate the association between weekends/weekdays and each of perfect taking (100% of prescribed doses taken) and timing (100% of prescribed doses taken on time) adherence. Taking (OR = 0.72 [95% CI 0.65-0.79]) and timing (OR = 0.66 [95% CI 0.59-0.74]) adherence were poorer on weekends than weekdays in the run-in (136 participants) and the intervention interval (taking OR = 0.74 [0.67-0.81] and timing OR = 0.71 [95% CI 0.65-0.77]). There was no interaction by treatment group (64 intervention and 74 control participants). Weekends represent a disruption of regular routines, posing a threat to adherence. Patients and families should be encouraged to develop strategies to maintain adherence when routines are disrupted. TAKE-IT registration number: Clinicaltrials.gov registration: NCT01356277 (May 17, 2011).

Induction Therapy in Elderly Kidney Transplant Recipients With Low Immunological Risk.

BACKGROUND: In nonimmunized patients, similar rejection rates are observed for patients who have undergone thymoglobulin (antithymocyte globulins [ATG]) or basiliximab (BSX) therapy. While ATG may improve delayed graft function, it may also be associated with higher infection rates and malignancy risk. We compared survival and clinical outcomes in elderly recipients with low immunological risk according to their induction therapy. METHODS: We conducted a multicentric study on nonimmunized patients ≥65 years of age receiving a first kidney transplant between 2010 and 2017. The principal outcome was patient and graft survival. Secondary outcomes were cumulative probabilities of infection, first acute rejection episode, malignancy, de novo donor specific antibody, posttransplant diabetes (PTD), cardiac complications, estimated glomerular filtration rate, and occurrence of delayed graft function. Cox, logistic, or linear statistical models were used depending on the outcome studied, and models were weighted on the propensity scores. RESULTS: Two hundred and four patients were included in the BSX group and 179 in the ATG group with the average age of 71.0 and 70.5 years, respectively. Patient and graft survival at 3 years posttransplantation were 74% (95% CI, 65%-84%) and 68% (95% CI, 60%-78%) in ATG and BSX group, respectively, without significant difference. Occurrence of PTD was significatively higher in BSX group (23% versus 15%, P = 0.04) due to higher trough levels of Tacrolimus on month 3 (9.48 versus 7.30 ng/mL, P = 0.023). There was no difference in other evaluated outcomes. CONCLUSIONS: In elderly recipients, ATG does not lead to poorer outcomes compared with BSX and could permit lower trough levels of Tacrolimus, thus reducing occurrence of PTD.

Gender Differences in Medication Adherence Among Adolescent and Young Adult Kidney Transplant Recipients.

BACKGROUND: Among kidney transplant recipients, gender differences in medication adherence may contribute to higher graft failure risks observed in girls and young women compared with boys and young men. Our aim was to determine whether adherence differs by gender, and whether gender differences vary by age in adolescent and young adult kidney transplant recipients. METHODS: We examined data from the 3-month run-in period (no intervention) of the randomized Teen Adherence in Kidney transplant Effectiveness of Intervention trial. Adherence was monitored using electronic pillboxes in 136 patients (11-24 y) followed in 8 transplant centers in Canada and the United States. We used ordinal logistic regression with generalized estimating equations to estimate the association between gender and each of daily taking (proportion of prescribed doses taken) and timing (proportion of prescribed doses taken on time) adherence, considering effect modification by age (11-16 y vs 17-24 y). RESULTS: No difference in taking adherence was observed by gender among participants aged 11 to 16 years (odds ratio [OR], 0.92; 95% confidence interval [CI], 0.55-1.54), whereas among participants aged 17 to 24 years, women had significantly greater odds of higher taking adherence scores (OR, 3.03; 95% CI, 1.20-7.66) than men. Results were similar for timing adherence, with no difference among participants aged 11 to 16 years (OR, 1.03; 95% CI, 0.65-1.63) but a greater odds of higher timing adherence scores in women than in men among participants aged 17 to 24 years (OR, 3.26; 95% CI, 1.43-7.45). There were no differences in adherence assessed by self-report or SD of tacrolimus trough levels. CONCLUSIONS: Gender differences in adherence vary by age. Whereas younger adolescents show no adherence differences by gender, young women show much better adherence than young men.

Association of Sex with Risk of Kidney Graft Failure Differs by Age.

Prior studies of sex differences in kidney graft survival showed conflicting results. We hypothesized that the association between recipient sex and kidney graft failure risk differs by recipient age and donor sex. We evaluated 159,417 patients recorded in the Scientific Registry of Transplant Recipients database who received a first deceased-donor kidney transplant (1995-2013). We used time-varying Cox models to estimate the association between recipient sex and death-censored graft failure. Models, stratified on donor sex and adjusted for potential confounders, included a recipient sex by current age interaction term. Among recipients of male donors, females of all ages had significantly higher graft failure risks than males (adjusted hazard ratios 0-14 years: 1.51 [95% confidence intervals 1.19 to 1.90]; 15-24 years: 1.37 [1.18 to 1.59]; 25-44 years: 1.14 [1.03 to 1.26]; 45 years: 1.05 [1.01 to 1.09]). Among recipients of female-donor grafts, only female recipients aged 15-24 years had a significantly higher graft failure risk than their male counterparts had (1.28 [1.06 to 1.53]). Indeed, female recipients aged ≥45 years had a significantly lower graft failure risk than their male counterparts had (0.95 [0.91 to 0.99]). These observations might be explained by the combined influence of several factors, including recognition of sex-determined minor histocompatibility antigens, influence of sex hormones on immune activation, sex- and age-related differences in medication adherence, and sex-related differences in body size. Additional studies should determine whether sex- and age-specific immunosuppression strategies are warranted for kidney graft recipients.

Absence of Decline of Kidney Function in Human Immunodeficiency Virus-Infected Patients Under Routine Clinical Management.

BACKGROUND: Since the introduction of antiretroviral therapy (ART), human immunodeficiency virus (HIV)-infected patients have a drastically improved prognosis but at the same time they are also more affected by non-HIV related complications, such as chronic kidney disease. The objective of our study was to investigate the effect of proteinuria and tenofovir (TDF)-containing ART regimens on the temporal evolution of estimated glomerular filtration rate (eGFR). METHODS: Between April 2008 and October 2012, we enrolled 395 patients with a complete renal evaluation among patients from the ANRS C03 Aquitaine cohort, a prospective hospital-based cohort of HIV-1-infected patients under routine clinical management in southwestern France. eGFR was estimated at each patient follow-up visit. A linear mixed model was used to analyze eGFR dynamics, accounting for change in TDF by modeling eGFR trajectory according to treatment periods. RESULTS: At inclusion, 56.7% of patients were treated with TDF-containing ART regimens; prevalence of glomerular and tubular proteinuria was 7.9 and 10.8% respectively. A 1-year increase of cumulative exposure to TDF was significantly associated with a mean eGFR decrease of 1.27 mL/min/1.73 m2 (95% CI [-2.14 to -0.41]). Only a urine protein to creatinine ratio >100 mg/mmol and/or a urine albumin to creatinine ratio >70 mg/mmol were associated with eGFR trajectory (mean slope 6.18 mL/min/1.73 m2 per year; 95% CI [2.71 to 9.65]), whereas TDF use was not associated with such eGFR temporal evolution. CONCLUSION: Decline in kidney function is limited under routine clinical management with monitoring of renal function and interventions including decision to continue or discontinue TDF.

Identifying subgroups of renal function trajectories.

Background: Renal function in patients with chronic kidney disease (CKD) may follow different trajectory profiles. The aim of this study was to evaluate and illustrate the ability of the latent class linear mixed model (LCMM) to identify clinically relevant subgroups of renal function trajectories within a multicenter hospital-based cohort of CKD patients. Methods: We analysed data from the NephroTest cohort including 1967 patients with all-stage CKD at baseline who had glomerular filtration rate (GFR) both measured by 51 Cr-EDTA renal clearance (mGFR) and estimated by the CKD-EPI equation (eGFR); 1103 patients had at least two measurements. The LCMM was used to identify subgroups of GFR trajectories, and patients' characteristics at baseline were compared between the subgroups identified. Results: Five classes of mGFR trajectories were identified. Three had a slow linear decline of mGFR over time at different levels. In the two others, patients had a high level of mGFR at baseline with either a strong nonlinear decline over time ( n = 11) or a nonlinear improvement ( n = 94) of mGFR. Higher levels of proteinuria and blood pressure at baseline were observed in classes with either severely decreased mGFR or strong mGFR decline over time. Using eGFR provided similar findings. Conclusion: The LCMM allowed us to identify in our cohort five clinically relevant subgroups of renal function trajectories. It could be used in other CKD cohorts to better characterize their different profiles of disease progression, as well as to investigate specific risk factors associated with each profile.

Analysis of risk factors associated with renal function trajectory over time: a comparison of different statistical approaches.

BACKGROUND: The most commonly used methods to investigate risk factors associated with renal function trajectory over time include linear regression on individual glomerular filtration rate (GFR) slopes, linear mixed models and generalized estimating equations (GEEs). The objective of this study was to explain the principles of these three methods and to discuss their advantages and limitations in particular when renal function trajectories are not completely observable due to dropout. METHODS: We generated data from a hypothetical cohort of 200 patients with chronic kidney disease at inclusion and seven subsequent annual measurements of GFR. The data were generated such that both baseline level and slope of GFR over time were associated with baseline albuminuria status. In a second version of the dataset, we assumed that patients systematically dropped out after a GFR measurement of <15 mL/min/1.73 m(2). Each dataset was analysed with the three methods. RESULTS: The estimated effects of baseline albuminuria status on GFR slope were similar among the three methods when no patient dropped out. When 32.7% dropped out, standard GEE provided biased estimates of the mean GFR slope in normo-, micro- and macroalbuminuric patients. Linear regression on individual slopes and linear mixed models provided slope estimates of the same magnitude, likely because most patients had at least three GFR measurements. However, the linear mixed model was the only method to provide effect estimates on both slope and baseline level of GFR unaffected by dropout. CONCLUSION: This study illustrates that the linear mixed model is the preferred method to investigate risk factors associated with renal function trajectories in studies, where patients may dropout during the study period because of initiation of renal replacement therapy.

Should we use standard survival models or the illness-death model for interval-censored data to investigate risk factors of chronic kidney disease progression?

BACKGROUND: In studies investigating risk factors of chronic kidney disease (CKD) progression, one may be interested in estimating factors effects on both a fall of glomerular filtration rate (GFR) below a specific level (i.e., a CKD stage) and death. Such studies have to account for the fact that GFR is measured at intermittent visit only, which implies that progression to the stage of interest is unknown for patients who die before being observed at that stage. Our objective was to compare the results of an illness-death model that handles this uncertainty, with frequently used survival models. METHODS: This study included 1,519 patients from the NephroTest cohort with CKD stages 1-4 at baseline (69% males, 59±15 years, median protein/creatinine ratio [PCR] 27.4 mg/mmol) and subsequent annual measures of GFR (follow-up time 4.3±2.7 years). Each model was used to estimate the effects of sex, age, PCR, and GFR at baseline on the hazards of progression to CKD stage 5 (GFR<15 mL/min/1.73 m2, n = 282 observed) and death (n = 168). RESULTS: For progression to stage 5, there were only minor differences between results from the different models. The differences between results were higher for the hazard of death before or after progression. Our results also suggest that previous findings on the effect of age on end-stage renal disease are more likely due to a strong impact of age on death than to an effect on progression. The probabilities of progression were systematically under-estimated with the survival model as compared with the illness-death model. CONCLUSIONS: This study illustrates the advantages of the illness-death model for accurately estimating the effects of risk factors on the hazard of progression and death, and probabilities of progression. It avoids the need to choose arbitrary time-to-event and time-to-censoring, while accounting for both interval censoring and competition by death, using a single analytical model.

Regression methods for investigating risk factors of chronic kidney disease outcomes: the state of the art.

BACKGROUND: Chronic kidney disease (CKD) is a progressive and usually irreversible disease. Different types of outcomes are of interest in the course of CKD such as time-to-dialysis, transplantation or decline of the glomerular filtration rate (GFR). Statistical analyses aiming at investigating the association between these outcomes and risk factors raise a number of methodological issues. The objective of this study was to give an overview of these issues and to highlight some statistical methods that can address these topics. METHODS: A literature review of statistical methods published between 2002 and 2012 to investigate risk factors of CKD outcomes was conducted within the Scopus database. The results of the review were used to identify important methodological issues as well as to discuss solutions for each type of CKD outcome. RESULTS: Three hundred and four papers were selected. Time-to-event outcomes were more often investigated than quantitative outcome variables measuring kidney function over time. The most frequently investigated events in survival analyses were all-cause death, initiation of kidney replacement therapy, and progression to a specific value of GFR. While competing risks were commonly accounted for, interval censoring was rarely acknowledged when appropriate despite existing methods. When the outcome of interest was the quantitative decline of kidney function over time, standard linear models focussing on the slope of GFR over time were almost as often used as linear mixed models which allow various numbers of repeated measurements of kidney function per patient. Informative dropout was accounted for in some of these longitudinal analyses. CONCLUSIONS: This study provides a broad overview of the statistical methods used in the last ten years for investigating risk factors of CKD progression, as well as a discussion of their limitations. Some existing potential alternatives that have been proposed in the context of CKD or in other contexts are also highlighted.

Representation of exposures in regression analysis and interpretation of regression coefficients: basic concepts and pitfalls.

Regression models are being used to quantify the effect of an exposure on an outcome, while adjusting for potential confounders. While the type of regression model to be used is determined by the nature of the outcome variable, e.g. linear regression has to be applied for continuous outcome variables, all regression models can handle any kind of exposure variables. However, some fundamentals of representation of the exposure in a regression model and also some potential pitfalls have to be kept in mind in order to obtain meaningful interpretation of results. The objective of this educational paper was to illustrate these fundamentals and pitfalls, using various multiple regression models applied to data from a hypothetical cohort of 3000 patients with chronic kidney disease. In particular, we illustrate how to represent different types of exposure variables (binary, categorical with two or more categories and continuous), and how to interpret the regression coefficients in linear, logistic and Cox models. We also discuss the linearity assumption in these models, and show how wrongly assuming linearity may produce biased results and how flexible modelling using spline functions may provide better estimates.