#5167 RENAL PROGNOSIS IN PATIENTS WITH ACUTE KIDNEY INJURY AND COMPARISON OF THE RIFLE AND KDIGO CLASIFICATION SYSTEMS: WHAT IS THE ROLE OF DIABETUS MELLITUS?

Abstract Background and Aims Acute Kidney Injury (AKI) is a frequent clinical entity. There are three AKI definition clasifications: RIFLE, AKIN and KDIGO. The endpoint of this study is to compare KDIGO and RIFLE regarding the prediction of AKI stage and adverse outcomes and to evaluate the renal prognosis of patients with a history of chronic kidney disease (CKD), who develop AKI during hospitalization. It is a well-known fact that Diabetes mellitus (DM) is the most common cause of end-stage chronic kidney disease (CKD). The epidemiology and outcome of diabetic patients with AKI during hospitalization, needs further study and comparison with the data of non-diabetic patients. As a secondary end point of this study we observed the progression of kidney function and severity of treatment in diabetic patients with stage III AKI (KDIGO) in comparison with the non-diabetic patients. Method This is a retrospective epidemiological study where data of 1083 adult patients were examined with AKI, who were hospitalized in various clinics at General Hospital of Athens Hippokratio between January 2021-September 2022. Results A total of 1083 patients were registered (63% male, mean age 74.5 ±12.6 years). 42% suffered from diabetes mellitus, 39% from arterial hypertension and 41% from CKD. RIFLE identified fewer patients with AKI than KDIGO (69.7% vs. 100%, p<0.001). Of all patients, 45.1% corresponded to AKI stage 1 (AKI-1), 7.3% to AKI stage 2 (AKI-2) and 47.6% to AKI stage 3 (AKI-3) according to KDIGO. 41.9% of AKI-3 patients underwent hemodialysis. Of the 214 patients on dialysis, 31.8% were not identified as AKI by RIFLE. In addition, patients with a history of CKD showed a higher percentage of AKI-3 (47.1% vs. 39.3% AKI-1, p=0.003 and 13.6% AKI-2, p=0.001) and underwent hemodialysis in a higher percentage (56.6% vs. 43.4%, p= 0.002). On the secondary end point 1083 patients, 458 (42.3%) diabetic and 625 (57.7%) non-diabetic were recognized. Hemodialysis was required in 83(18%) diabetic patients and in 132(21.1%) non-diabetic patients. Patients who ended up on end stage CKD requiring chronic hemodialysis or with GFR<15ml/min/1.73 m2 one month after hospital discharge were 37.7% and 51, 1% respectively. More specifically patients with the worst prognosis, who developed end stage CKD or ended up with GFR<15ml/min/1.73 m2 one month after hospital discharge, were categorized in three groups according to CKD Stage at hospital admission : in group A) CKD STAGE II out of 29, 8 (27.3%) non-diabetics and out of 19, 4 (21%) diabetic, in group B) STAGE IIIa out of 30 patients 5 (16.6%) non-diabetic and out of 15 3 (20%) diabetic and in group C) STAGE IIIb out of 17 patients 8 (47%) non-diabetic and out of 20 6 (30%) diabetics. It should be mentioned that the worst prognosis was observed in patients who were hospitalized for infection out of 30, 16(63%), with prerenal AKI out of 19 patients, 10(53%) and with a cardiovascular event out of 18 patients, 6(33%). Conclusion RIFLE criteria identified a lower proportion of patients with AKI compared to KDIGO, while not classifying a proportion of patients undergoing hemodialysis as AKI. Additionally, CKD was associated with worse renal prognosis in hospitalized patients with AKI. Patients with AKI on CKD who required hemodialysis during their hospitalization did not appear to have a worse prognosis at one month after discharge,compering diabetic with non-diabetic in group A and B. On the contrary there was a significant difference between patients in group C.