The researchers excluded patients allergic to TXA, anemic
patients, patients with coagulopathy, patients on anticoagulation therapy, patients with history of thromboembolic
events, patients with creatinine > 2.0 mg/dL, patients with
chronic liver disease and pregnant patients. Informed consent
was obtained for all patients. Randomization was through
computer-generated numbers, performed in blocks of four.
Patients, anesthesiologists and surgeons were blinded to
treatment group (control group receiving placebo or study
group receiving TXA).
During the 18-month study period, 300 patients were eligible but 220 were excluded based on the exclusion criteria.
Of the remaining 80 patients, one patient in each group opted
for simpler surgeries than inclusion criteria dictated, leaving
78 total patients in the study, with 39 patients in each group.
Identical infusion bags of placebo or TXA were prepared
ahead of time by research staff who were not involved in patient care. Group I received 0.9% NaCl or placebo and Group
II received 2 doses of 15 mg/kg of TXA. The first dose was
given over 20 minutes before induction of anesthesia and the
second dose was given 3 hours after the first dose. All patients
had standard intraoperative monitoring, standard general
anesthesia, and controlled hypotension during the surgery,
with a mean arterial blood pressure (MAP) of about 20% below preoperative values and a minimum MAP of 60 mmHg.
Intraoperative blood loss was estimated by the volume of
fluid in the suction bottles and the weight of swabs, minus
the amount of irrigating fluid used. Hematocrit levels were
checked at least every two hours during surgery. Postopera-
tive blood loss was based on drain output. The threshold for
blood transfusion was a hematocrit level of 30%. The primary
outcome was perioperative blood loss up to 24 hours after
surgery and the secondary outcome was the incidence of
Comparisons of the two groups showed no statistical differences in demographics and preoperative characteristics.
Both groups were also comparable regarding types of surgery
performed, surgeon experience, operative time and hospital
length of stay.
For the primary outcome measure, the results showed that
the control group had significantly higher blood loss than the
TXA group, intraoperatively and postoperatively. Intraoperative estimated blood loss (EBL) in the control group was a
median of 700 mL (range 150-3500 mL) and in the TXA group
was a median of 450 mL (range 80-4000 mL). The control group
also had a significantly higher number of patients with intraoperative EBL > 500 mL and intraoperative EBL > 1000 mL.
For the secondary outcome measure, the control group
had a significantly higher number of patients who received
intraoperative blood transfusions ( 33.3% as compared to
12.8% in the TXA group). After surgery, but within the first 24
hours, postoperative blood transfusions were not significantly
different between the two groups. The total blood loss for the
whole perioperative period was a median of 900 mL in the
control and 600 mL in the TXA group (statistically different).
In the whole perioperative period, 17 patients in the control
group received a blood transfusion, compared to six patients
in the TXA group (statistically different). A separate regression
analysis was performed on the patients who had intraoperative
EBL greater than 500 mL, in order to determine risk factors
for significant bleeding. The researchers found that operative
time greater than 180 minutes had increased odds (odds ratio
of 5.99) for significant bleeding and that receiving TXA was
protective (odds ratio of 0.25).
Seventy-eight patients undergoing multilevel decompression
or decompression with fusion thoracolumbar surgeries were
randomized to receive either two doses of perioperative
tranexamic acid (TXA) or placebo.
The treatment group received 15 mg/kg of TXA 20 minutes
prior to induction of anesthesia and a second dose three
Intraoperatively, the control group had significantly higher
blood loss than the TXA group (median of 700 mL in the
control group compared to median of 450 mL in the TXA
In the perioperative period (surgery and the first 24 hours
after surgery), 17 patients in the control group received a
blood transfusion, compared to six patients in the TXA group.
The findings support the use of TXA for reducing blood
loss and limiting blood transfusions in patients undergoing
complex thoracolumbar surgeries.
Strengths of Study
This study addresses a paramount question in spine surgery, as
blood transfusions are associated with perioperative adverse
events and increased costs.
The authors performed a prospective, blinded, randomized
controlled trial, with sufficient power, thus providing
Level 1 evidence supporting the use of TXA in complex
Limitations of Study
A variety of TXA dosing regimens are described in the literature,
but only one dosing regimen, which has not previously been
studied in spinal surgery, was used in this trial.
This study included a significant mix of cases and surgical
indications, which led to a wide range of reported blood losses
(80 mL to 4000 mL).
Several aspects of this research may not be transferrable to
other institutions, such as controlled hypotension during
surgery and a transfusion threshold hematocrit of 30%.