Introduction: Urine output provides a rapid estimate for kidney function, and its use has been incorporated in the diagnosis of acute kidney injury. However, not many studies had validated its use compared to the plasma creatinine. It has been showed that the ideal urine output threshold for prediction of death or the need for dialysis was 0.3 ml/kg/h. We aim to assess this threshold in our local ICU population. Methods: This was a secondary analysis of an observational study done in critically ill patients. Hourly urine output data was collected, and a moving average of 6-hourly urine output was calculated over the first 48 hours of ICU admission. AKIuo was defined if urine output ≤ 0.5 ml/kg/h, and UO0.3 was defined as urine output ≤ 0.3 ml/kg/h. Results: 143 patients were recruited into the study, of these, 87 (61%) had AKIuo, and 52 (36%) had UO0.3. The AUC of AKIuo in predicting death was 0.62 (0.51 to 0.72), and UO0.3 was 0.66 (0.55 to 0.77). There was lower survival in patients with AKIuo and UO0.3 compared to those without (p=0.01, and 0.001, respectively). However, only UO0.3 but not AKIuo independently predicted death (HR 2.44 (1.15 to 5.18). Conclusions: A threshold of 6 hourly urine output of 0.3 ml/kg/h but not 0.5 ml/kg/h independently predictive of death. This support previous finding of a lower threshold of urine output criteria for optimal prediction.

Introduction: There has been increasing evidence of detrimental effects of cumulative positive fluid balance in critically ill patients. The postulated mechanism of harm is the development of interstitial oedema, with resultant increase morbidity and mortality. We aim to assess the impact of positive fluid balance within the first 48 hours on mortality in our local ICU population. Methods: This was a secondary analysis of a single centre, prospective observational study. All ICU patients more than 18 years were screened for inclusion in the study. Admission of less than 48 hours, post-elective surgery and ICU readmission were excluded. Cumulative fluid balance either as volume or percentage of body weight from admission was calculated over 6, 24 and 48 hour period from ICU admission. Results: A total of 143 patients were recruited, of these 33 died. There were higher cumulative fluid balances at 6, 24 and 48 hours in nonsurvivors compared to survivors. However, after adjusted for severity of illness, APACHE II Score, they were not predictive of mortality. Sensitivity analysis on sub-cohort of patients with acute kidney injury (AKI) showed only an actual 48-hour cumulative fluid balance was independently predictive of mortality (1.21 (1.03 to 1.42)). Conclusions: Cumulative fluid balance was not independently predictive of mortality in a heterogenous group of critically ill patients. However, in subcohort of patients with AKI, a 48-hour cumulative fluid balance was independently predictive of mortality. An additional tile is thus added to the mosaic of findings on the impact of fluid balance in a hetergenous group of critically ill patients, and in subcohort of AKI patients.

Introduction: The mNUTRIC score is a nutritional assessment tool to identify critically ill patients with high nutritional risk who could benefit from nutritional interventions. This study was conducted to validate the 28-day mortality prognostic performance of the mNUTRIC score in a Malaysian intensive care unit (ICU). Methods: This was a retrospective cohort study of adult patients who were consecutively admitted to the ICU from January 2017 to December 2018 for >24 hours. Data were collected on variables required to calculate the mNUTRIC score. Patients with mNUTRIC score ≥5 points were considered to be at high nutritional risk. Main outcome was 28- day mortality from all causes; ICU length of stay (LOS) and prolonged mechanical ventilation (MV) (>2 days) were secondary outcomes. Results: From a total of 432 admissions, 382 (88.4%) patients fulfilled the study criteria. Seventy-seven (20.2%) of these patients were at high nutritional risk. They had longer mean ICU LOS (7.1±7.5 days versus 4.2±4.0 days, p=0.001), greater proportion of prolonged MV (57.1% versus 14.4%, p<0.001) and higher 28-day mortality (44.2% versus 10.2%, p<0.001) compared to patients with low mNUTRIC score (≤4 points). High mNUTRIC score predicted 28-day mortality with area under the curve (AUC) of 0.797 (95% confidence interval: 0.738-0.856). Conclusion: High mNUTRIC score was associated with a higher 28-day mortality. The prognostic performance for 28-day mortality of the mNUTRIC score is clinically valid as indicated by AUC >0.7 and is comparable to the results of other validation studies. In addition, patients with high mNUTRIC score had increased ICU LOS and prolonged MV.

Introduction: This work aims to establish the practicality of simple point-of-care (POC) measurements of procalcitonin (PCT) coupled with the standard PCT-guided antibiotic treatment discontinuation algorithm to guide the cessation of antibiotic treatment in intensive care unit (ICU). Methods: In this randomised-controlled trial, 80 adult patients with suspected bacterial infections were randomised to either the POC PCT-guided arm (n = 40) or the standard-of-care arm (n = 40). The decision to discontinue antibiotic treatment in the POC PCT-guided arm was based on the POC PCT-guided antibiotic-treatment discontinuation strategy, which states that discontinuation is urged once the PCT concentration has reduced by ≥ 80% or to < 0.5 ng/mL. In the standard-of-care arm, the antibiotic-treatment duration followed the local guidelines. Results: The median duration of antibiotic treatment was 6.5 [IQR = 5.0-7.0] days in the POC PCT-guided antibiotic-treatment arm versus 7.5 [IQR = 5.0-14.0] days in the standard-of-care arm (p = 0.010). The mean antibiotic-free days in the first 30 days after study inclusion was 20.7 (SD = 5.3) days in the POC PCT-guided antibiotic-treatment arm versus 16.4 (SD = 7.4) days in the standard-of-care arm (p = 0.004). The number of patients who took an antibiotic for more than 10 days was 2 (5%) in the POC PCT-guided antibiotic-treatment arm versus 13 (32.5%) in the standard-of-care arm (p = 0.002). Conclusion: Antibiotic use in patients with symptoms of bacterial infections in the ICU was substantially minimised with the installation of a POC PCT-guided antibiotic-treatment cessation.