In the present retrospective study, it was hypothesized that application, in scoliosis surgery, of a protocol for blood and fluid management, based on goal directed fluid therapy, cell salvage and tranexamic acid, could lead to reduced allogeneic red blood cells transfusion. The patients will be enrolled in a retrospective observational study and divided in two groups. Patients in no protocol Group received a liberal intraoperative fluid therapy and patients in protocol Group received fluid therapy managed according to a stroke volume variation based protocol. The protocol included fluid therapy according to SVV monitor, permissive hypotension, tranexamic acid infusion, restrictive RBC trigger and use of perioperative cell savage.

At investigators institution, neither anesthetic approaches to replacement of blood or fluid losses were standardized before 2014 for scoliosis surgery. Before 2014, fluid therapy was liberal and according to general principles of good clinical practice and ephedrine boluses of 5 mg were given when fluid boluses failed to maintain a systolic arterial pressure >90 mm Hg. Blood was replaced with crystalloid at a 3:1 ratio and colloid at a 1:1 ratio. Regarding blood product transfusion anesthesiologists were generally initiated when hemoglobin levels were less than 8 g/dl or less than 10 g/dl in patients with coronary diseases and predonated autologous or allogeneic RBCs were administered.

A protocol of management for scoliosis surgery was implemented in 2014 and included: a) fluid therapy according to SVV monitor, b) intraoperative permissive hypotension to reduce active bleeding (goal mean arterial pressure 60 mmHg), c) prophylactic tranexamic acid infusion (30 mg/kg bolus, 1mg/kg/hr during surgery), d) restrictive RBC trigger according to national standardized protocols (Hb<7.0 g/dL or <9g/dL in patients with coronary diseases) and e) use of perioperative cell savage.

In patients of Group Pro, basal crystalloid infusion was started at 4 ml/kg/h right after general anesthesia induction and intubation. ClearSight System (Edwards Lifesciences Cop, Irvine, CA, USA) was used to measure stroke volume variation and cardiac output, continuously and noninvasively through finger- cuffed technology. If SVV was >15% rapid crystalloid bolus of 10ml/kg or 4ml/kg colloid bolus were administered until it reached a value of ≤15%. After two consecutive fluid boluses SVV remained >15%, administration of noradrenaline infusion was considered.

Data will be collected from anesthesia records and included: age, gender, height, weight, body mass index and ASA score. Additional variables included infused crystalloid volume, infused colloid volume and the number of allogeneic transfused units of RBC. Serum Hb levels were measured preoperatively and after the end of surgery. Moreover, diuresis and use of vasopressors use were recorded.

Inclusion Criteria:

1. severe curve (Cobb Angle > 70-degree)
2. same surgeon in all procedures

Exclusion Criteria:

1. Patients with coagulopathy,
2. morbid obesity
3. severe cardiopulmonary disease
4. liver dysfunction
5. renal disease

Patients scheduled for elective orthopaedic surgery of scoliosis, with a planned intensive care unit admission, were enrolled in a retrospective observational study.