POS-040 DIAGNOSTIC VALUE OF URINE MICROSCOPY IN PREDICTING OUTCOME OF HOSPITALIZED PATIENTS WITH ACUTE KIDNEY INJURY: A SINGLE-CENTER PROSPECTIVE STUDY

Abstract

Acute kidney injury (AKI) is common in hospitalized and critically ill patients and has an important modifying effect on mortality, kidney recovery and health resources. Evaluation of the urine sediment is often viewed as a complementary measure for providing additional insight into the diagnosis and severity of AKI; however, it has not formally been integrated in consensus definitions. This is a prospective observational cohort study aimed at identifying the diagnostic value of urine microscopy of in-patients admitted at Perpetual Succour Hospital from May 2016 to May 2017 with AKI, specifically intrinsic and prerenal AKI in relation to predicting outcome such as worsening AKI, need for in-hospital dialysis, or in-hospital death. Eligible patients had clinical data and serum creatinine levels extracted. Urine samples were collected, and each had a dipstick test done. A single investigator did the evaluation of urine sediment. A urine microscopy scoring system was derived based from the number of granular casts, muddy-brown casts and renal tubular epithelial cells. The urine microscopy score was correlated to mortality, worsening or recovery of renal function, and need for renal replacement therapy (RRT). A total of 102 patients were eligible. The average age was 63.47 years old, and majority were males (59.8%). Sepsis was the most common cause of AKI (48%) followed by Cardiorenal Syndrome (20.6%). Most patients were hypertensive and diabetic (33.3%). As for patients’ creatinine results, their average baseline was 1.93 mg/dl with an eGFR of 43.32 ml/min. Creatinine levels during urine microscopy averaged at 4.72 + 2.97, and upon discharge, creatinine levels were 2.39 + 1.93 with corresponding eGFR of 39.94 + 34.75 ml/min. The urine microscopy score did not show a statistically significant association with regards to greater need for dialysis and its discontinuation (p value of 0.131 and 0.500 respectively). However, there is a trend towards an association with the patient’s mortality, though not statistically significant (p value= 0.084). Tests of correlations using Pearsons test showed that the relationship between urine microscopy scoring system and rise in creatinine percentage from baseline (r = 0.075, p = 0.475), as well as between the percent decrease in creatinine (r = 0.105, p = 0.349) were weak and not statistically significant. The presence of hematuria on patient’s outcome showed that the presence of RBC >10/hpf was significantly associated with the need for RRT (p= 0.015) and mortality (p= 0.039). Also, the presence of Renal Tubular Cells was significantly associated with dialysis (p= 0.002), while its absence was associated with no need for RRT (p= 0.002). Most importantly, the absence of Renal Tubular cells was significantly associated with renal recovery (p= 0.047). Moreover, there is a trend towards the association of the presence of Muddy Brown Casts and mortality (p=0.066). The presence of Renal tubular epithelial cells, muddy brown casts and hematuria are associated with mortality and need for RRT. More importantly, there is a trend towards an association of a higher urine microscopy score with mortality. Septic AKI seems to have higher urine microscopy scores; this finding may imply that the urine microscopy scoring system may be most useful in septic AKI in predicting outcome.