People with genetic kidney diseases on kidney replacement therapy have different clinical outcomes compared to people with other kidney diseases – Scientific Reports

Study design and setting

This retrospective observational study included people who commenced KRT between 1 January, 1989 and 31 December, 2020 as recorded in the ANZDATA registry. De-identified information on patient and graft donor variables were received from the ANZDATA registry which stores information on people on kidney replacement therapy in Australia and New Zealand (Tables 1 and 2). The study was approved by the Human Research Ethics Committee (HREC) of the University of Tasmania (20,409) and ANZDATA steering committee (42,745). All methods were performed in accordance with relevant guidelines and regulations. All individual participants provided informed consent on entry into ANZDATA registry for research approved by ANZDATA and local HREC.

Study variables

Primary kidney disease codes in ANZDATA registry align with European Renal Association primary kidney disease codes. Primary kidney disease diagnoses are annotated by kidney specialist based on clinical assessment and are not always genetically- nor biopsy- proven. To enhance capture of GKDs, GKDs were subclassified as majority monogenic and minority monogenic based on percentage of cases within each disease classification with monogenic bases as previously described²⁰ (Supplementary Table S1). Majority monogenic GKDs were defined by evidence from cohort/case studies that ≥ 50% cases have an identifiable monogenic bases. Minority monogenic were defined by evidence from cohort/case series studies that < 50% cases have an identifiable monogenic basis. Monogenic basis was defined as a likely pathogenic or pathogenic (ACMG variant classification) variant or variants with the appropriate zygosity in a gene with an established or justified gene-phenotype/kidney disease relationship. Phenocopy disorders were excluded from the assessment of monogenic basis for kidney disease. Evidence was drawn from Online Mendelian Inheritance in Man, PanelApp Australia and the ClinGen Clinical Domain Working Groups²¹. People with kidney diseases that had no monogenic basis were included as a comparator group. People were included in the dialysis cohort if they did not receive a kidney transplant by 31 December, 2020. People were included in the transplant cohort if they received a kidney transplant.

Primary outcome measures included 10-year mortality for the dialysis and transplant cohort; and 10-year graft failure for the transplant cohort. Graft failure occurred when the graft was no longer functioning (excluding death with functioning graft). Secondary outcome measures such as cause of death, cause of graft failure and disease in graft kidney were recorded for descriptive purposes. Patient age, gender, smoking status, body mass index (BMI), ethnicity, comorbidities (diabetes, chronic lung disease, coronary artery disease, peripheral vascular disease, cerebrovascular disease), first KRT modality, dialysis vintage were assessed for the dialysis cohort. Additional variables such as donor source, donor age, cold ischemia time, human leukocyte antigen (HLA) mismatch, dialysis vintage and transplant era were assessed for the transplant cohort.

Statistical analysis

Categorical demographic variables were reported using counts and percentages (Tables 1 and 2). Continuous demographic variables were reported was means and standard deviations. Chi-squared test of independence were used for categorical variables and Analysis of Variance (ANOVA) were used to compared continuous variables.

Kaplan–Meier survival curves were used to assess time from dialysis initiation (dialysis cohort) or transplant date (transplant cohort) to death or graft failure. In the transplant cohort, Kaplan–Meier survival curves were used to visualize time from transplant date to graft failure. Cox proportional hazards regression were used to calculate unadjusted and adjusted hazard ratio (AHR) of mortality in the dialysis and transplant cohorts; and graft failure in the transplant cohort. Mortality and graft failure were assessed as time-to-event variables. In the dialysis cohort, disease type, age in decades, gender, ethnicity, smoking status, BMI, smoking status, comorbidities, first dialysis modality and dialysis vintage were included into the mortality model as covariates. In the transplant cohort, disease type, recipient age in decades, gender, ethnicity, smoking status, BMI, comorbidities, first dialysis modality, dialysis vintage in years; donor source and age in decades; cold ischemia time in hours, HLA mismatch, and transplant era were included into the mortality and graft failure models as covariates. Test for proportional hazard assumption were initially completed using log–log survival curves (Supplementary Fig. 1). Where log–log survival curves intersect (i.e. mortality in transplant cohort), Schoenfeld residual plots for each variable were individually inspected to further assess nonproportionality per Therneau and Grambsch²² (Supplementary Figs. 2–4).

Adjusted hazard ratios (AHRs) for death-censored graft failure and graft failure-censored mortality were assessed using cause specific hazard model²³. Subgroup analyses were completed to assess the effect of key constituents of each GKD group (e.g. polycystic kidney disease for majority monogenic and reflux nephropathy for minority monogenic GKD) on the overall hazard ratios.

Only complete cases were included in the analyses. For all analyses, a significance level of 0.05 was used. Statistical analyses were performed using SPSS software (IBM Corp. Released 2021. IBM SPSS Statistics for Windows, Version 28.0. Armonk, NY: IBM Corp). Study was reported per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies²⁴.

• The Renal Warrior Project. Join Now
• Source: https://www.nature.com/articles/s41598-024-57273-x