The study was approved by the ethics committee of the S. Anna University Hospital, Ferrara, Italy (Ref: 28/1/10) and registered with ClinicalTrials.gov (Ref: NCT01320891). Informed consent was obtained from all subject prior to the randomization.
This was a double blind randomized trial. Eligible patients were those aged 18 years or older, undergoing abdominal surgery for bowel cancer. Exclusion criteria were: 1) emergency surgery for bowel punch or intestinal occlusion; 2) cardiac failure (New York Heart Association class III or IV); 3) kidney dysfunction (serum creatinine > 200 μmol/L); 4) preoperative anemia (hemoglobin < 10 g/dl); 5) therapy with corticosteroids or nonsteroidal anti-inflammatory substances; 6) allergy to hydroxyethyl starches and 7) patient rejection of participation in the study.
Patients were randomly allocated in a 1:1 ratio in two groups, the balanced solution (BS) group in which the fluids administered were always balanced solutions, both as colloids and crystalloids; and the unbalanced solution (UBS) group in which the fluids administered were always unbalanced solutions, both as colloids and crystalloids. The randomization list was created by using a computer based block randomization by a statistician prior to the initiation of the study. Patients fulfilling the inclusion criteria received an increasing sequential number in accordance with the order of their randomization and inclusion in the study. The allocation concealment was ensured by pharmacy-controlled randomization: on the basis of the randomization list, the pharmacies distributed patient-specific colloid and crystalloid bottle that were identical in size, weight and appearance. Patients allocated to BS group were treated with colloid bottles containing 6% HES 130/0.42 in plasma adapted Ringer’s acetate/malate solution (Tetraspan®; B.Braun Melsungen, Germany) and crystalloid bottles containing plasma adapted Ringer’s acetate/malate solution (Sterofundin® ISO; B.Braun, Germany). Patients allocated in UBS group received colloid bottles containing 6% HES 130/0.42 diluted in unbalanced normal saline solution (Amidolite®; B.Braun, Germany) and crystalloid bottles containing normal saline solution (NaCl 0.9%, Fresenius Kabi, Germany). All study investigators, staff members and patients were blinded to the study medication.
Volume administration was started after induction of anaesthesia and continued until 8 a.m. of the first postoperative day. A ratio of 3:1 between crystalloids and colloids was used in both groups. As part of our routine management of fluid administration, the quantity was based on maintaining a mean arterial pressure (MAP) of at least 60 mmHg and a central venous pressure (CVP) of about 10–11 mmHg during surgery.
Till 24 h after the beginning of surgery, fluid administration continued using 1500–2000 ml of respectively unbalanced solution in the UBS group and balanced solution in the BS group. During the entire study period, packed erythrocytes were given when hemoglobin was less than 7 g/dl in patients with normal cardiac function or less than 9 g/dl in patients with ischemic heart disease.
Induction of anaesthesia was performed with propofol (2 mg/kg), fentanyl (3 μg/kg) and vecuronium (0.1 mg/kg) for neuromuscular blockade. Anaesthesia was maintained with fentanyl (2–3 μg/kg/h), sevoflurane (at least 1 MAC and vecuronium, titrated according to the patients’ needs; or with a continuous infusion of propofol, remifentanyl, vecuronium or cisatracurium at weight related doses. Nonsteroidal anti-inflammatory drugs were not administered throughout the investigation period. A warming cover blanket system and fluid warmers were used to avoid hypothermia during surgery.
Perioperative monitoring included measurement of the electrocardiogram, arterial blood pressure, central venous pressure, oxygen saturation and end tidal CO2.
Measurements of clinical variables were performed immediately after anaesthesia induction and before volume administration (T0), at the end of surgery (T1), within 2 h after surgery (T2) and 24 h after the beginning of surgery (T3). The following data were collected: 1) blood gases variables, including lactate concentrations from arterial and venous blood samples; 2) electrolytes, albumin and total serum protein from venous blood sample; the strong ion difference (SID) was determined as previously described ; 3) neutrophil gelatinase-associated lipocalin (NGAL) from urinary sample; 4) MMP-9 total and active, TIMP-1, IL-6, IL-8, IL-10 from venous blood samples; the MMP-9/TIMP-1 ratio was calculated as an index of equilibrium between the action of MMP-9 and its inhibition.
IL-6, IL-8 and IL-10 detection assay
IL-6, IL-8 and IL-10 levels were simultaneously measured in sera of patients, diluted 1:2 with dilution buffer, by a multiplex sandwich enzyme-linked immunosorbent assay (ELISA) system based on chemiluminescence detection (Aushon SearchLight chemiluminescent array kits; Tema Ricerca, Italy) according to the manufacturer’s recommendations. The interleukin levels are reported as pg/ml. The detection limits were 0.19 pg/ml (IL-6), 0.39 pg/ml (IL-8) and 0.39 pg/ml (IL-10).
MMP-9 and TIMP-1 detection assay
Peripheral blood samples were collected in anticoagulant-free test tubes and kept in ice until centrifugation at 3,000 rpm for 10 min. Serum samples were stored in aliquots at −80°C until assay. Serum concentrations of active and total (active + inactive) MMP-9 were measured by using a commercially available activity assay system kit (GE Healthcare, RPN2634), as previously described in details . Serum levels of TIMP-1 were measured by using a commercially available “sandwich” enzyme-linked immunosorbent assay kit (GE Healthcare, RPN2611) as previously described in details .
NGAL detection assay
The urinary NGAL was determined by using the ArchitectR analyzer (Abbott Diagnostics, Illinois, USA).
To detect a difference in MMP-9/TIMP-1 ratio of 0.15 with an SD of 0.14, with a type I error of 0.05 and a power of 0.80, 30 patients should have had to be recruited, 15 in each group. Considering a dropout rate of 20% and in view of the limited comparability of the study condition, the sample size calculation resulted in n = 2 × 20 patients. The approximate degree of normal distribution was calculated for each parameter by the Kolmogorov-Smirnov test. Comparisons for continuous variables within and between groups were performed with the Friedman repeated-measures analysis of variance on ranks. To isolate divergent variables, multiple comparison procedures were used (Dunnett’s Method). A P value of less than 0.05 was accepted as statistically significant.