Cerebral Aneurysm

This was a longitudinal study of a prospective cohort of CKD patients in Korea, called KNOW-CKD (KoreaN cohort study for Outcome in patients With Chronic Kidney Disease). KNOW-CKD is a multicenter prospective cohort study that enrolled adult predialysis patients with CKD stages G1 to G5^{11 (link)}. Patients were classified into four groups according to the specific cause of CKD at enrollment: glomerulonephritis (GN), diabetic nephropathy (DN), hypertensive nephropathy (HTN), and Polycystic kidney disease (PKD). Each group classification was determined based on pathologic diagnosis if a biopsy result was available (27.6% of total patients: 66.5% of GN group, 6.4% of DN group, 7.3% of HTN group and 1.4% of PKD group). Otherwise, group classifications were based on clinical diagnoses. The biopsy-proven GN consisted as following – 40% IgA nephropathy, 7% focal segment glomerular sclerosis, 6% membranous nephropathy, 5% crescentic GN, 2.4% minimal change disease, and 1.5% lupus nephritis. Non-biopsy-proven GN was defined as the clinical history manifesting chronic GN and the presence of albuminuria or glomerular hematuria with or without an underlying systemic disease causing GN. The active GN population taking immunosuppressant at enrollment was excluded to minimize the heterogeneity by treatment. Diagnosis of DN was strictly based on albuminuria in a patient with type 2 diabetes and the presence of diabetic retinopathy. To exclude DN patients who may have combined GN, diabetic patients with glomerular hematuria were not included in the DN group. HTN was diagnosed by a history of hypertension and the absence of a systemic illness associated with kidney damage. Only the patients with proteinuria < 1.5 g/day and a proportion of urine albumin < 50% of urine protein were included in HTN to exclude the GN population. To diagnose PKD, unified ultrasound criteria were used^{12 (link)}. Other causative diseases was categorized as ‘unclassified’ and excluded from our analysis.
A total of 2238 patients enrolled in the study from April 2011 to February 2016. After excluding patients with unclassified etiology or without follow-up data, 2070 patients were finally analyzed in this study for survival analysis with follow up until March 31, 2020. To determine the annual eGFR change and trajectory, we included only those patients (n = 1952) with more than two creatinine measurements (Fig. 1). Written informed consent from each patient was collected voluntarily at the time of enrollment. The study was approved by the institutional review board of each participating hospital: Chonnam National University Hospital (CNUH-2011-092), Eulji General Hospital (201105-01), Gil Hospital (GIRBA2553), Kangbuk Samsung Medical Center (2011-01-076), Pusan Paik Hospital (11-091), Seoul National University Bundang Hospital (B-1106/129-008), Seoul National University Hospital (H-1704-025-842), Seoul St. Mary’s Hospital (KC11OIMI0441), and Yonsei University Severance Hospital (4-2011-0163). This study follows the guidelines of the 2008 Declaration of Helsinki.Figure 1

Flowchart of enrolled study patients. eGFR, estimated glomerular filtration rate; IDMS, isotope dilution mass spectrometry.

Demographic details and medication history were collected at enrollment. Serum creatinine was measured at each study visit by a central laboratory (Lab Genomics, Seoul, Republic of Korea) using an isotope dilution mass spectrometry-traceable method. For eGFR, the CKD -EPI equation based on serum creatinine was used^{13 (link)}. After the baseline visit, patients were followed-up at 6 and 12 months and then every 1 year until death or drop-out and follow-up events were recorded. In case of loss to follow-up, patients were censored for kidney and CVD events at the last follow-up visit. Death and the cause of death were collected using either hospital medical records or data from the National Database of Statistics Korea using the Korean resident registration number. Data were collected until whichever came first: drop-out, death, or March 31, 2020.
Both kidney failure and the composite of kidney failure and/or creatinine doubling were used as kidney outcomes. Kidney failure was defined as starting maintenance dialysis (required for longer than 3 months) or receiving kidney transplantation. Another outcome was the composite outcome of CVD and all-cause death. CVD was defined as any first event of the following that needed hospitalization, intervention, or therapy during the follow-up period : acute myocardial infarction, unstable angina which needed admission due to aggravated coronary ischemic symptoms, percutaneous coronary artery intervention or coronary bypass graft surgery, ischemic or hemorrhagic cerebral stroke, cerebral artery aneurysm, congestive heart failure, symptomatic arrhythmia, aggravated valvular heart meant by requiring hospital admission, any pericardial disease that required hospital admissions such as pericarditis, pericardial effusion, or cardiac tamponade, abdominal aortic aneurysm, or severe peripheral arterial disease (Table S1).
The chi-square test or Anova was used to compare the baseline characteristics. Non-normally distributed variables such as parathyroid hormone, urine protein/creatinine, and high sensitivity C-reactive protein were compared by Kruskal–Wallis test. The four groups had significant differences in baseline characteristics including age and baseline eGFR; we therefore used the overlap propensity score (PS) weighting method to minimize the effects of confounding factors on outcomes^{14 (link)}. Overlap weighting is a PS method that tries to mimic important attributes of randomized clinical trials. This method can overcome the potential limitation of adjusting the difference in measured characteristics using classic PS methods of inverse probability of treatment weighting (IPTW). Overlap weighting overcomes these limitations by assigning weights to each patient that are proportional to the probability of that patient belonging to the opposite group^{15 (link)}. PSs were calculated using a logistic model with the following variables since they showed significant differences among the four groups: age, sex, body mass index, CKD stage, mean blood pressure, CVD, hemoglobin, serum uric acid, calcium, phosphorous, albumin, total cholesterol, high-density lipoprotein, low-density lipoprotein, fasting blood sugar, intact parathyroid hormone, urine protein-to-creatinine ratio, high-sensitivity C-reactive protein, diuretics use, statin use, and angiotensin converting enzyme inhibitor or angiotensin receptor blocker use in this study. The log10 transformed values were used for PS calculation with the non-normally distributed variables such as parathyroid hormone, urine protein-to-creatinine ratio, and high sensitivity C-reactive protein. The patients in the compared group were weighted by the probability of the reference group (1-PS), and the patients in the reference group were weighted by the probability of the compared group (PS). For two groups of CKD causes, we applied the overlap weighting method to each set, resulting in a total of 6 sets. To visually compare distributions of balance, the density plots were created (Figure S1). Additionally, the standardized mean difference (SMD) was calculated to check good balance after the overlap weighting method was applied. This is calculated by the absolute value of the difference in mean among groups divided by the standard deviation. The SMD less than or equal to 0.10 means good balance after weighting^{15 (link)}. In outcome comparison analysis, a Cox proportional hazard model was used for kidney outcomes, and a cause-specific hazard model was used for the composite of CVD and death. In the competing risk model for the composite of CVD and death, kidney failure was considered a competing risk since many patients who started kidney replacement therapy were no longer followed for further event thereafter. Results are presented as hazard ratios (HRs) and 95% confidence intervals (95% CI). To estimate annual eGFR change, generalized linear mixed models were constructed with random intercepts and slopes with an unstructured model for the correlation structure. The results were expressed as estimates (standard errors). In the adjusted models, the variables used in PS score calculation were further adjusted. Spaghetti plots showing the individual trajectories of eGFR during follow-up were drawn to determine patterns of eGFR decline according to cause of CKD. P for the quadratic term was tested using polynomial mixed models with random intercepts and slopes. A P value less than 0.05 was considered statistically significant. SAS 9.4 (SAS Institute, Cary, NC, USA) and R version 3.5.3 (Foundation for Statistical Computing, Vienna, Austria) were used.