Albumin infusion rate and plasma volume expansion: a randomized clinical trial in postoperative patients after major surgery

Patients scheduled for non-emergent Whipple operation or major gynecological cancer surgery ≥ 40 years of age were approached by a member of the research team. The patients were informed about the possibility to participate in the study after admission to the postoperative care unit, should they fulfill the postoperative inclusion criteria. Written informed consent was collected from interested patients. Patients who had given a written consent prior to the operation were screened by a member of the research team the first 5 h after admission to the post-anesthesia care unit. The inclusion criteria were (1) written consent and (2) indication for fluid therapy as judged by the physician caring for the patient and at least one of the following criteria: (a) positive “leg raising test” (pulse pressure increase > 9%) [20, 21], (b) central venous oxygen saturation < 70%, (c) arterial lactate > 2.0 mmol/l, (d) urine output < 0.5 ml/kg the hour prior to inclusion, (e) respiratory variation of the inferior vena cava of more than 15% as measured by ultrasound [22, 23], and (f) systolic blood pressure < 100 mmHg or mean arterial blood pressure < 55 mmHg.

Patients fulfilling any of the following criteria were excluded: (1) hypersensitivity to the active drug or the tracer; (2) signs of postoperative bleeding; (3) history of heart failure; (4) the physician caring for the patient considered that there were strong reasons to administer another fluid or the same fluid, but in another way or in a different volume than stated in the protocol; (5) pregnancy; and (6) clinical judgment by the investigator or the treating physician that the patient should not participate in the study for reasons other than described above. Predefined permanent withdrawal criteria included the change of surgical procedure or the occurrence of serious adverse events.

Eligible patients were randomized at a ratio of 1:1 using sealed envelopes to receive 5% albumin at a dose of 10 ml/kg of predicted body weight [24] in either 30 min or 180 min. Sealed envelopes were prepared by an independent party (Clinical Research Unit, Skåne University Hospital, Lund). Randomization was performed using a computerized random number generator, and the research team was blinded to block size.

The primary outcome was change in plasma volume from the start to 180 min after the start of albumin infusion. Secondary outcomes were change in the area under the plasma volume curve from the start to 180 min after the start of infusion of albumin, transcapillary escape rate (TER) for albumin, changes in hemodynamic parameters, diuresis, plasma concentration of hormones involved in fluid homeostasis, and plasma concentration of glycocalyx components and incidence of postoperative complications up to 30 days postoperatively (see Additional file 1: Table S1 for the definitions of complications).

Plasma volume was measured by calculating the distribution volume of ¹²⁵I-human serum albumin (SERALB-125®, CIS Bio International, Gif-Sur-Yvette, France) administered intravenously before the start of the albumin infusion, at 30 min, and at 180 min after the start of the infusion. The method is considered to be the gold standard for the measurement of plasma volume [25] and is described in more detail in the supplement. The change in the area under the plasma volume curve was calculated using the trapezoid rule. Transcapillary escape rate for albumin is a measure of leakage of albumin from microvessels into the interstitium and was measured by measuring plasma concentration of ¹²⁵I-human serum albumin at 10, 30, 45, and 60 min after the last injection of the tracer (Additional file 1: Figure S1). The decrease in plasma concentration of ¹²⁵I-human serum albumin was fitted to be a mono exponential function and is expressed as the percentage of the decrease in plasma concentration of ¹²⁵I-human serum albumin per hour [26‐28]. Potential sources of error in measurement of transcapillary escape rate for albumin and plasma volume have been evaluated previously and have been found to be small [19, 26, 29]. Transcapillary escape rate was measured between 180 and 240 min after the start of the infusion to ensure that both groups had received the same dose of albumin when the measurement was performed.

Plasma concentrations of glypican-4 (Cloud-Clone Corp), hyaluronan (Echelon Biosciences), Syndecan-1 (Diaclone), renin (IDS), copeptin (Brahms GmbH), albumin (Roche), and Mid Regional-pro Atrial Natriuretic Peptide (MR-proANP) (Brahms GmbH) were determined by immunologic assays according to the manufacturer’s instructions before and at 180 min after the start of the albumin infusion. On a post hoc basis, we analyzed albumin before and at 180 min after the start of the albumin infusion using an immunological assay (Roche). Blood gas analysis and determination of hematocrit and plasma lactate were performed using a blood gas analyzer (Radiometer 850, Radiometer). Hematocrit was measured every 30 min from baseline to 180 min after the start of infusion. The participants received routine postoperative care after the study protocol was completed. Measurement of plasma volumes, transcapillary escape rate, and plasma concentrations of glycocalyx components and hormones were made blinded to treatment. Baseline hemodynamic data and blood gases were collected prior to treatment allocation whereas these parameters were recorded by an investigator aware of the treatment allocation at the later time points. For an overview of the measurements, see Additional file 1: Figure S1.

Comorbidities, medications, routine laboratory analysis results, American Society of Anesthesiology (ASA) classification, and Revised Cardiac Risk Index were collected from the hospital electronic chart system or calculated. Perioperative data were collected from the anesthesia chart. Hemodynamic data were recorded immediately prior to the start and at 180 min after the start of albumin infusion. Diuresis was recorded 4 h before and 6 h after the start of infusion.

In experimental studies, a 6-ml/kg greater plasma volume expansion was found after slow administration of 5% albumin at a dose of 12 ml/kg compared with a bolus dose of the same volume [13, 14]. Based on these data, the present study was powered to detect a difference in plasma volume expansion between the groups of 4 ml/kg. Assuming a standard deviation of the change in plasma volume of 5 ml/kg [30], about 30 patients in each group were required to obtain a power of 80% using an unpaired t test. To adjust for the possibility that patients did not complete the protocol, we aimed to include 35 patients in each arm.

Statistical analyses were performed per protocol by an independent statistician blinded to the treatment allocation, and the main results were interpreted as specified in the protocol [19]. Hypothesis testing was made using two-tailed testing, and a P value below 0.05 was considered significant. Assessment of normality was performed using histograms and the Shapiro-Wilk test and followed by the unpaired t test or the Mann-Whitney test as appropriate. Fisher’s exact test was used for categorical variables. Analysis of covariance (ANCOVA) and Pearson correlation analysis were used for the sensitivity analysis. Influence of baseline blood volume and type of surgery on treatment effect were assessed in pre-specified sensitivity analyses. Data are presented as median and interquartile range or as mean ± SD. SAS 9.4 Institute Inc., Cary, NC, USA, was used for the analysis.

A total of 70 patients were enrolled between the 18 June 2014 and the 22 November 2016, and 35 patients were assigned to each treatment. Prior to the second amendment, 1 patient that had given consent was not included due to an ongoing vasopressor therapy. For a CONSORT flowchart of patients, see Fig. 1. Two patients experienced a suspected allergic reaction following the administration of ¹²⁵I-human serum albumin and were not included in the study. One patient withdrew consent during the intervention. Plasma volume measurement was unsuccessful due to technical problems in 3 patients. A total of 33 patients receiving slow infusion and 31 patients receiving rapid infusion were included in the analysis. Pre-treatment characteristics are presented in Table 1. The two most common criteria for fluid administration were a positive passive leg raising test and an elevated arterial lactate. All patients with a positive passive leg raising test also had at least 1 additional objective criteria suggesting hypovolemia. A total of 4 patients received treatment with norepinephrine at inclusion, and in all cases, the dose was below 0.1 μg/kg/min. Baseline plasma volume was 46.9 ± 9.2 ml/kg and 47.7 ± 6.3 ml/kg in the slow and rapid groups, respectively (Fig. 2). This corresponded to a calculated blood volume of 73.1 ± 15.0 ml/kg and 72.9 ± 10.3 ml/kg in the slow and rapid groups, respectively.

The increase in plasma volume from the start to 180 min after the start of infusion did not differ between the two infusion rates and was 7.4 ± 2.6 ml/kg and 6.5 ± 4.1 ml/kg in the slow and rapid infusion groups, respectively (mean difference, 0.9 [95%CI, − 0.8 to 2.63.2], P = 0.301, t test, Fig. 3). Change in the area under the plasma volume curve over time was smaller in the slow infusion group than in the rapid infusion group and was 866 ± 341 min ml/kg and 1226 ± 419 min ml/kg in the slow and rapid groups, respectively (mean difference, 360 min ml/kg [95%CI, 169 to 550], P < 0.001, t test) (Table 2). The number of patients with postoperative complications in the slow and rapid infusion groups was 8 and 6, respectively, and did not differ between the groups (P = 0.774) (Table 2 and Additional file 1: Table S2). No treatment effect on transcapillary escape rate for albumin, lactate, hematocrit, or any of the hemodynamic parameters could be detected (Table 2). Overall diuresis from the start of infusion of albumin to 360 min after the start of infusion did not differ between the treatment groups and was 1.0 ± 0.4 and 1.1 ± 0.6 ml/kg/h (mean difference, − 0.1 ml/kg/h [95%CI, − 0.1 to 0.4, P = 0.325, t test). In a post hoc analysis, hourly diuresis during this time period was analyzed using a mixed linear regression model and diuresis as measured at 60 to 180 min from the start of the infusion was higher in the rapid infusion group than in the slow infusion group (Additional file 1: Figure S7). A significant interaction between the time period treatment effect was detected suggesting a lower diuresis in the slow group during the earlier hours and higher diuresis at the later periods relative to the rapid infusion group (P = 0.002 for interaction between time period and the treatment group, Additional file 1: Figure S7). Plasma albumin concentration increased more in the slow than in the rapid group (Table 2).

Plasma concentration of the stable precursor fragment of atrial natriuretic peptide, mid-regional pro-atrial natriuretic peptide, increased more from the start to 180 min after the start of infusion in the rapid infusion group than in the slow infusion group whereas renin and copeptin, the latter reflecting vasopressin release, did not differ between the groups (Table 2). Glypican-4, a component of the endothelial glycocalyx, increased more in the slow infusion group than in the rapid infusion group whereas no difference in the change in hyaluronan and syndecan-1 could be detected (Table 2).

The pre-planned sensitivity analysis did not demonstrate an interaction between baseline blood volume and treatment effect (P = 0.075, two-way ANCOVA, Additional file 1: Figure S1) or between the type of surgery and treatment effect (P = 0.364, two-way ANCOVA). To further explore if the treatment effect was dependent on baseline blood volume, patients with baseline blood volume above and below the median were analyzed separately, and the results aligned with the ANCOVA results (Additional file 1: Figure S3). The treatment effect in Whipple and gynecological surgery patients was also analyzed separately, and no difference in the primary outcome could be demonstrated in either of the groups (Additional file 1: Figure S5). On a post hoc basis, change in 1-Hct over time was analyzed and was 3.6 ± 1.7 and 6.0 ± 2.6%/min in the slow infusion and rapid infusion groups, respectively (mean difference 2.4, [95%CI, 1.3–3.5%/min, P < 0.001, t test) (Additional file 1: Figure S4).