This observational study evaluates the potential benefits, costs and clinical outcomes of albumin over saline and other non-saline fluids in patients receiving large volume resuscitation.

Currently, recommended resuscitation paradigms involve a "crystalloid-first" approach. This approach reserves more potent and potentially more toxic colloids for patients that have already been exposed to large amounts of crystalloid and may have also experienced a delay in correcting their shock.

This observational retrospective cohort study aims to identify predictors for large volume resuscitation (LVR) and model the potential benefits, costs and clinical outcomes of albumin over saline and other non-saline fluids in patients receiving large volume resuscitation. Further analyses will assess the risk of acute kidney injury (AKI) and hyperchloremic metabolic acidosis (HCA) associated with resuscitation fluid choice and examine long term outcomes such as development of end stage renal disease and post-discharge mortality up to 1 year following hospital discharge in patients treated with various fluid types

This study will utilize data in a large, heterogeneous cohort (n=~65,800) of critically ill patients admitted to the ICU over a 12 year period at the University of Pittsburgh Medical Center. The study population will consist of patients who receive large volume resuscitation (defined as > 60ml/kg in a single 24 hour period) separated into analysis groups based on fluids administered during the defined 24 hour large volume resuscitation window. Patients presenting with AKI or HCA prior to large volume resuscitation will be excluded. AKI and "Severe" AKI will be diagnosed based on KDIGO criteria within 72 hours following LVR. Metabolic acidosis will be based on arterial blood gas measurements with a base deficit > 2 mEq/l, with patients having chloride as the ion contributing to the majority of the acidosis being diagnosed with Hyperchloremic metabolic acidosis. Mortality at 30, 90, and 365-days from ICU admission will be ascertained using the social security death master file. Propensity score models will be used to determine the adjusted relationship between these clinical outcomes and type of fluid resuscitation employed.

Inclusion Criteria:

• Critically ill patient admitted to ICU

Exclusion Criteria:

• History of chronic dialysis and/or renal transplant
• Baseline serum creatinine >= 4 mg/dl
• Develop AKI or MA prior to LVR
• Received OR fluids during LVR

Adult patients admitted to any of eight ICUs at the University of Pittsburgh Medical Center.