#2745 New strategies for capturing undiagnosed chronic kidney disease in patients at risk

Abstract

Abstract Background and Aims Chronic Kidney Disease (CKD) is recognized as a health problem in the general population. The worldwide incidence of chronic kidney disease (CKD) is increasing, driven by aging populations and a higher prevalence of type 2 diabetes (T2D) and hypertension. The CKD is generally asymptomatic, and the diagnosis depends on the laboratory monitoring of the patients. The KDIGO consensus statement recommend the implementation of screening programs in patients at high risk of CKD development. We present the first results of a sentinel surveillance program of CKD in a healthcare area for detecting undiagnosed CKD in patients at risk. Method Our health department covers the metropolitan area of Valencia, attending 341, 972 citizens through 31 primary care centers. A CKD screening program has been established in this population based on a middleware clinical decision support (CDS) system “CDS-Ripple Down - Abbott Diagnostics” integrated in an electronic request system and in the electronic health records. When a general practitioner doctor order a lab test, the middleware CDS system detects high risk patients for CKD defined by age: >65 to <90, diabetes mellitus, hypertension, or obesity. Automatically, the system adds serum creatinine, estimated glomerular filtration rate (eGFR), urine albumin-to-creatinine ratio (UACR) and urine sediment analysis if it has not been requested (sentinel program). Then, the patients are classified into KDIGO stages based on eGFR and UACR and the CDS system detect those with progression (Fig. 1). Only patients with UACR > 300 mg/g, eGFR < 30 ml/min/1.73, or progression were referred to nephrology. A nephrologist then decides whether patients require a face-to-face visit or provides recommendations to primary care physicians. Results From 01/11/2023 to 31/12/2023, 4, 989 reports were generated by CDS-system corresponding to 122 laboratory test per day. 99, 33% were added by the sentinel program. 3, 970 (79.6%) patients did not have CKD (KDIGO G1-2 or UACR <30 mg/g), 887 (17.8%) patients CKD not suitable for referral to Nephrology (KDIGO 3 and UACR 30-300 mg/g without progression, and 130 (2, 6%) were referred to Nephrology. (Fig. 2) Patients referred to Nephrology had a median age of 79 years [IQR: 72-85], 68 (52, 3%) were women. Twenty-seven patients (20%) were classified as KDIGO G2 A3, 19 (14, 6%) as KDIGO G4 A1 and 18 (13, 8) as KDIGO G2 A2. Fifty-seven patients (43, 8%) have been scheduled to face-to-face at nephrology consultation (1.1% of the total sample), and 73 (56, 1%) have been referred to their primary care doctors with recommendations (1.5% of the total sample). Among the patients referred to Nephrology by the CDS system, the reasons for not scheduling a face-to-face visit in Nephrology were mild decrease in eGFR (n = 18), UACR A2 (n = 24), elderly patients with low Kidney Failure Risk Equation (n = 5), urological pathology (n = 10), dependent patient and/or palliative situation (n = 5) and other causes (n = 11). Conclusion A novel CKD automatic screening method for capturing undiagnosed CKD among patients at risk has been developed. After screening around 5 000 patients at risk in two months, 2.6% of them presented criteria for referral to Nephrology. Only 1.4% required face-to-face visit. If we continue at this screening rate, it is expected that half of our population at risk of developing CKD will have been screened in less than one year. Computer systems with algorithms programmed and improved by a multidisciplinary team can establish a sentinel route (reviewing patients medical records), interpreting analytical values (calculating progressions, KDIGO stages, KFRE risk...) and producing automatic interpretive reports that integrates all the elements of CKD clinical attention allowing us to carry out this population screening.