Aim of this study is to reduce postoperative complications (assessed by POMS-Score and graded by Clavien-Dindo-Score) by substitution of perioperative hypoalbuminaemia in high-risk patients or patients undergoing high risk surgery.

Perioperative hypoalbuminaemia is a common condition in high risk surgery and in high risk patients. It is associated with increased morbidity and mortality. Although many studies investigated albumin substitution in critically ill patients there is still no clear recommendation for its use over any other colloid. Except for some smaller studies especially in the setting of cardiac surgery the effect of perioperative albumin substitution has not been studied yet.

This study is designed to investigate the effect of perioperative albumin substitution in high risk patients or patients undergoing high risk surgery on postoperative complications assessed by the POMS-Score and graded according to severity by the Clavien-Dindo-Score.

Patients eligible for the study are informed in detail during the pre-medication visit and written informed consent will be obtained. Additionally the Quality of Recovery 9 Score (QoR-9) will be determined. During induction of anesthesia blood is taken from every patient and the serum albumin level is determined. In case albumin is greater than 30 g/l albumin levels are controlled every 3 hours and one last time on admission to the postoperative anaesthesia care unit (PACU). Should the albumin level do not fall below 30 g/l over the whole period the patient is not randomized.

Patients with albumin ≤30g/l will be randomized using a web-based computer-generated randomization list into one of two groups:

Group human albumin (HA):

In this group hypoalbuminaemia is treated with infusion of human albumin until serum albumin is greater than 30g/l. To verify success of the therapy, serum albumin levels are controlled every 3 hours during surgery, on admission to the PACU and if necessary again in the PACU.

Group control:

In this group patients are treated according to standard clinical care. To compare the effect of albumin substitution serum albumin is measured every 3 hours from beginning of anesthesia until discharge from the PACU.

In both groups patients will be discharged from the PACU when the patient is rated fit for discharge by the attending anesthetist and albumin level is >30g/l in the HA group.

A blinded investigator assesses the patients postoperative complications. He visits the patient on day 3, 5, 8 and 15 after surgery and determines the POMS-score (see reference: Grocott MP). If a complication is detected the assessor grades the complication according to its severity with the Clavien-Dindo-Score (see reference: Clavien PA). Additionally serum albumin and creatinine levels will be determined in routine blood tests. The patients QoR-9, the Charlson Comorbidity Score, the Preoperative Score to Predict Postoperative Mortality (POSPOM) and other secondary outcome (see Outcome Measures section) parameters will be obtained.

6 Month after surgery the patient or next of kin will be contacted in order to determine 6-month mortality and the QoR-9 score.

Postulating an incidence of complications (graded Clavien-Dindo-Score 2 or higher) of 52% in group control and a reduced incidence of 38% in group HA results in a total number of 400 patients (chi-square-test, α = 0,048, power 80%). As the actual incidence is unknown but has a significant impact on the number needed per group, an interims analysis will be performed after 100 included patients. Is the difference of the incidence of postoperative complications between the two groups significant with an α < 0,002 the study will be finished. Otherwise, the sample size needed per group will be corrected assuming a relative risk reduction of 25% (52% vs. 38%), an α = 0,048, a two-tailed test and a power of 80%.

The blinded and independent interims analysis after 100 patients revealed a corrected sample size of 300 patients per group (total 600 patients).

Inclusion Criteria:

• age: 18 years or older
• written informed consent
• ASA state 3 and 4 OR
• elective high risk surgery (open aortic surgery, open leg revascularization, thrombectomy, oesophagectomy, pancreatic surgery, cystectomy, liver surgery, change of knee- or hip-prothesis, amputation)

Exclusion Criteria:

• emergency surgery
• severe liver cirrhosis (child pugh C)
• need for dialysis
• patients already included in SUPERADD
• patients with a legal representative
• contraindications for human albumin: hypersensitivity for human albumin or any substance included in the preparation
• pregnancy
• breastfeeding women
• ASA state 5
• BMI > 35 kg/m2