142 TIMING OF NEPHROLOGY REFERRAL AND OUTCOME IN ACUTE KIDNEY INJURY PATIENTS RECEIVING CONTINUOUS RENAL REPLACEMENT THERAPY

Abstract

Category: Epidemiology and Outcomes from AKI Presenter: Dr ASIAH USAMAH Keywords: Acute kidney injury, continuous renal replacement therapy, timing of referral Timing of nephrology referral in critically ill acute kidney injury patients is a subject of debate. Some studies have found that late nephrology referral is associated with poor prognosis. No consensus exists on the best timing of nephrology consult and initiation of CRRT. To determine the factors and outcome in relation to timing of nephrology referral in acute kidney injury (AKI) patients receiving continuous renal replacement therapy (CRRT). This is a retrospective study involving all AKI patients aged more than 18 years old receiving CRRT in ICU, HKL from October 2016 till April 2017. A total of 107 patients were screened and 68 patients were included in the study. Patients were divided into 2 categories based on timing of referral from onset of AKI. Early referral is defined as referral less than 48 hours while late referral is more than 48 hours. AKI is defined as per KDIGO guideline. Data were collected from ICU local database and analysed using SPSS version 20. Categorical and demographic data were expressed as frequency (n) and percentage (%), whilst continuous variables was expressed as mean and standard deviation (SD). Sixty eight patients, comprising 51 early referral and 17 late referral were analysed. Patients who were referred late had higher mortality (n=15, 88.2%) compared to early referral group (n=35, 15.7%, p=0.22). Patients who were in the late referral group also had longer length of stay, 12days (3.5-22) compared to patients in early referral group, 4 days (2-10), p=0.049. There were no significant difference in blood urea, serum creatinine, serum lactate, bicarbonate, serum potassium and urine output volume at referrals in both groups. Late referral group had higher mortality, albeit statistically not significant most likely due to small sample size. This study did not include disease severity as a confounding factor for mortality.