The invesitgators utilized the Cerner HealthFacts database (a large HIPAA-compliant clinical-administrative database maintained by Cerner Inc., USA) to identify a cohort of 6,249 adults that underwent on-pump cardiac surgery for valve and/or coronary artery procedures between January 2001 and March 2013. Of these, the investigators selected 1136 patients who received 5% albumin on the day of or the day following cardiac surgery and matched them (1:1) with 1136 patients who did not receive albumin. Characteristics on which patients were matched included patient demographic, hospital and procedural characteristics, baseline patient comorbidities including preoperative CKD and 26 other grouped conditions, as well as acute severity of illness at admission to ICU.

Following development of the study plan and project summary, approval was obtained (November 25th 2013) from the Duke University Medical Center institutional review board, who reviewed the study outline, including endpoints and treatment exposure groups prior to data extraction. HIPAA-compliant data were then extracted from the U.S. Cerner HealthFacts®, (Cerner Corp., Kansas City, MO) database. Patients who underwent cardiac surgery utilizing CPB between Jan 2001 and March 2013 were identified. In addition to hospital characteristics (bed-size, teaching status, location), encounter-level patient data (demographics, admission source, payer), comprehensive time-stamped medication orders, pharmacy records, laboratory results, admission and discharge diagnoses (International Classification of Diseases Ninth Revision Clinical Modification, ICD-9-CM codes) and procedures were available.

Study Population Adults that underwent isolated valve, isolated CABG or two or more procedures utilizing CPB, and who survived for at-least 24 hours were eligible for inclusion (except when CABG was performed emergently following a PCI). Procedures were identified using ICD-9-CM codes (39.61 = extracorporeal circulation auxiliary to open heart surgery) or CPT codes (indicating 'on-pump' surgery). Patients with missing or incomplete admission, procedure, and/or discharge dates, gender, and age were excluded. Discrete encounters less than four hours apart within the same hospital system, were considered contiguous. We also excluded those patients undergoing heart transplantation (ICD-9 procedure code 37.51) or with any of the following Elixhauser comorbidities: AIDS, lymphoma, metastatic cancer, solid tumors without metastases. Only the first operative procedure during this time was considered and patients undergoing multiple non-contiguous procedures were excluded.

Definition of Exposure The study cohort consisted of patients that had either received at least 500mL of any crystalloid (0.9% saline, buffered salt solutions such as Plasma-Lyte, Lactated Ringers etc.) with or without any volume of 5% albumin - the 'albumin' and 'crystalloid only' cohorts respectively. All intravenous fluids documented on the day of or the day following surgery were considered in order to determine eligibility for inclusion, and fluid volumes from the day of through two days following surgery were monitored. The investigators excluded crystalloids administered in smaller containers (<=250mL volume) as these were most likely to be used for medication delivery rather than for the maintenance of hydration or expansion of intravascular blood volume in this adult population. Such volume thresholds were not imposed for colloids in the 'albumin' cohort however. Patients were excluded from the study if they received any other colloid (e.g. albumin over 5% concentration or hydroxyethyl starch) or hypertonic saline. Patients were not excluded from the albumin arm if they received plasma protein fraction (PPF) however.

Definition of Outcomes The primary outcome was in-hospital mortality during the index admission in which cardiac surgery was performed. The investigators also examined in-hospital mortality within 30, 60, and 90 days after the index procedure. Secondary endpoints included major postoperative adverse cardiac, renal, hepatic, metabolic, pulmonary, neurologic, hematologic, or infectious events defined using clinical parameters or administrative codes (ICD-9-CM). Additional administrative secondary outcomes included: Length-of-stay, readmissions, and re-operations (return to operating room). MACE at 30, 60 and 90 days were calculated from four separate binary categories: 1) in-hospital mortality; 2) post-operative stroke or cerebrovascular accident per ICD-9 diagnosis codes not present-on-admission for the index visit or for subsequent visits within respective MACE timeframes; 3) post-operative revascularization or PCI; and 4) post-operative myocardial infarction. A composite binary indicator was also reported for any one or greater MACE condition met within the respective timeframes.

Inclusion Criteria:

• Adults that underwent isolated valve, isolated CABG or two or more procedures utilizing CPB

Exclusion Criteria:

• death within 24 hours of index procedure
• patients undergoing heart transplantation
• Patients with missing or incomplete admission, procedure, and/or discharge dates, gender, and age were excluded
• patients with any of the following Elixhauser comorbidities: AIDS, lymphoma, metastatic cancer, solid tumors without metastases

Adults that underwent isolated valve, isolated CABG or two or more procedures utilizing CPB

Individual data are de-identified