This was a randomized, double-blind, clinical study carried out at the Julio Muller University Hospital (Mato Grosso State, Brazil). The study was approved by the hospital Research Ethics Committee registered under number 635/CEP-HUJM/09 and is in accordance with the ethics principals set out in the Helsinki Declaration (2000), and meets Brazilian national legal specifications. The study was registered in ClinicalTrails.gov under the number NCT01354249. External monitoring of the study was carried out by the Research Ethics Committee of the Julio Muller Hospital.
Inclusion criteria includes adults (18-65 years-old), of both sexes, and candidates to elective moderate operations such as open cholecystectomy, laparoscopic cholecystectomy and unilateral repair of inguinal hernia). Exclusion criteria were acute cholecystitis, diabetes mellitus, chronic kidney failure, chronic liver disease ot serum bilirubin greater than 2 mg/dL, body mass index (BMI) above 35 Kg/m2, American Anesthesiologists Association (ASA) score above 2, gastro-esophageal reflux, gastroparesis or intestinal obstruction. Patients with any non-compliance with the study protocol, or who had associated operations, or presented significant intraoperative occurrences, or experienced prolonged operations (lasting more than 3 hours) were also excluded.
Patient randomization was carried out on admission to the hospital using random numbers issued by a computer program . For the randomization the precepts of the CONSORT flow diagram were followed .
The patients were randomized into two groups: the carbohydrate-protein group (CHO-P) and the control group (CO). The patients were given a specific drink to their group on the evening prior to surgery and three hours before the operation. The CHO-P group received 474 ml (evening drink) or 237 ml (3 h prior to operation drink) of a solution containing 14% whey protein (100% lactoalbumin), 86% carbohydrates (45% maltodextrine and 55% sucrose) and 0% lipids (Resource® Breeze - Nestlé, São Paulo, Brasil) and the CO group received the same volume of water. All the patients fasted for solids for 6 hours from the operation.
On the day of the surgery and on the first postoperative day blood samples were collected for glucose, insulin, triglycerides, albumin, pre-albumin, CRP, and α-1-acid glycoprotein (α-1-GA) assays. Preoperative samples were collected at the infirmary approximately one hour before the patient was sent to the surgical center.
The HOMA-IR (Homeostasis Model Assessment-Insulin Resistance) equation was used as proposed by Matthews et al  to assess insulin resistance according to the formula: HOMA-IR= insulin (μU/mL) × glycaemia (mg/dL)/405. To assess inflammatory activity the PINI (CRP (mg/L) × α-1-GA (mg/L)/albumin (g/L) × pre-albumin (mg/L) and the CRP/albumin ratio were used .
Nutritional status was assessed by BMI and subjective global assessment . The patients received intraoperatively a single dose of 2 g of intravenous cefazolin and were hydrated with 1000-1500 ml of Ringer's lactate during immediate postoperative period. Dipyrone (500 mg IV every 6 hours) was prescribed for pain and anti-emetics were used only when necessary.
The sampling calculation was based on a previous study . A quantity of 8 cases in each study branch was judged to be sufficient to ensure 80% power (beta error) and 5% significance (alpha error) expecting a difference in 50% in in the insulin resistance (Homa equation). All the continuous data were initially analysed for homogeneity by the Levene test and for normality by the Kolmogorov-Smirnov test. The Student t test or Mann-Whitney test was then applied accordingly. A cut-off for normality was for serum CRP (48 mg/L) and CRP/albumin ratio (11) based on the overall mean of the data plus 1 standard deviation. A significance level of 5% (p < 0.05) was established. The data were presented as a mean and standard deviation or as a median and variance. All the calculations were made on a computer using the Statistical Package for the Social Sciences (SPSS) for Windows 11. 0.