Oral Is as Effective as Intravenous Tranexamic Acid at Reducing Blood Loss in Thoracolumbar Spinal Fusions: A Prospective Randomized Trial : Spine

Secondary Logo

Journal Logo

RANDOMIZED TRIAL

Oral Is as Effective as Intravenous Tranexamic Acid at Reducing Blood Loss in Thoracolumbar Spinal Fusions

A Prospective Randomized Trial

Yu, Charles C. MDa; Fidai, Mohsin MDb; Washington, Travis MDb; Bartol, Stephen MDb; Graziano, Gregory MDb

Author Information
SPINE 47(2):p 91-98, January 15, 2022. | DOI: 10.1097/BRS.0000000000004157
  • Free

Intraoperative and postoperative blood loss with elective spinal fusion surgery adversely affects patient outcomes by increasing coagulopathy, hematoma formation, and anemia.1 The ensuing need for allogenic blood transfusion gives rise to increased risk of infections, long-term mortality, transfusion reactions, and economic burden.2–4 Achieving optimal perioperative blood conservation, through prophylactic administration of antifibrinolytic agents, has been the focus in recent years. Specifically, a lysine analoge named tranexamic acid (TXA) reduces perioperative blood loss and need for blood transfusions in spine surgery by exerting its anti-fibrinolytic effect through the reversible blockade of lysine binding sites on plasminogen.5–7

As the utilization of TXA becomes more ubiquitous in spine surgery, the search for the most efficacious route of TXA administration as well as dosing regimen becomes worthwhile. TXA can be given intravenously (IV), topically, and orally (PO). Much of the spine literature focuses on the IV or topical formation.8 Potential advantages of the PO formulation include lower medication cost and ease of administration. There is recent evidence supporting the use of PO TXA in the adult reconstruction literature,9,10 which also showed that PO TXA is more cost effective than IV TXA. Similarly, our initial phase prospective randomized study showed equivalent blood loss with IV versus PO TXA.11 However, the preliminary data had insufficient patient numbers to perform subgroup analysis.

Therefore, the objective of this final phase was to compare the efficacy of IV and PO routes of TXA on perioperative blood loss and allogeneic blood transfusion rate in adult patients undergoing posterior thoracolumbar instrumented fusion surgery that varied in fusion levels.

MATERIALS AND METHODS

Study Design and Patients

This study was a single-center, randomized, prospective trial designed to test equivalence between IV and oral TXA in the setting of elective thoracolumbar fusion between February 2017 and March 2020. Our institutional review board approved the present study, and it was registered with the public ClinicalTrials.gov registry (NCT03037515). After obtaining informed consent, adult patients (age ≥18 years’ old) undergoing elective posterior thoracolumbar instrumented spinal fusions were enrolled in the study. Surgical indications included spinal stenosis, spondylolysis, spondylolisthesis, degenerative disc disease, adult spinal deformity, proximal junctional kyphosis, adjacent level disease, and pseudoarthrosis. Exclusion criteria included known allergy to TXA, history of renal failure or kidney transplant, history of arterial thromboembolic event (e.g., myocardial infarction, stroke) within the past year, placement of an arterial stent within the past year, a history of thromboembolic event, coagulopathy, or refusal to receive blood products.

Intervention

Enrolled patients were randomly assigned between the two treatment groups of IV and oral TXA using a computer-generated random number algorithm. Surgeons and data collectors were blinded to the patient allocation. Because the number of fusion levels can influence blood loss, the sample was subcategorized before randomization by the number of vertebral levels arthrodesed (one to two levels, three to five levels, more than five levels). Serum and pharmacokinetic studies have demonstrated that IV TXA reaches therapeutic concentration rapidly but falls below the threshold after 5 hours; PO TXA reaches therapeutic levels after 2 hours and maintains levels above the therapeutic threshold for 6 hours after administration.11 Based on pharmacokinetic data, the oral TXA group received 1950 mg TXA (three tablets of 650 mg) approximately 2 hours before incision. The IV TXA group received the standard dosing for our institution of 1 g TXA (diluted in 100 mL normal saline) given as an IV bolus immediately before incision and another 1 g TXA given before closure.

Surgical Technique and Postoperative Care

Posterior instrumented spinal fusions were performed prone on a Jackson table with the abdomen free. A forced-air warming device was used to maintain normothermia. A standard open midline approach was utilized, intertransverse fusion beds were prepared uniformly, and pedicle screws were inserted in the standard fashion. When indicated, transforaminal lumbar (TLIF) and direct lateral interbody (DLIF) fusions were also performed per surgeon preference. Before wound closure, hemostasis was achieved and drains were placed routinely. All cases were performed by surgeons with >25 years’ experience in spine surgery.

Blood Management

An autologous blood recovery system (Cell Saver, Haemonetics, MA) was used when requested by the surgeon and salvaged red blood cells (RBCs) were returned to the patient. Suctioned blood from the surgical field was processed and given back to the patient if estimated blood loss was >500 mL or at the surgeon's discretion. As a result, not every patient may have been given salvaged RBCs. Intraoperative blood transfusion was given if hemoglobin (Hgb) dropped <7.0 g/dL or at the anesthesiologist's discretion, such as if patient was unstable despite fluid resuscitation and salvaged RBC replacement. The postoperative transfusion protocol required transfusion for a Hgb level <7.0 g/dL or if Hgb was between 7.0 and 8.0 g/dL and patients had symptomatic anemia including tachycardia, hypotension, or pallor.

Data Collection

Patient demographic and preoperative characteristics were documented for comparison between the treatment groups. All patients had postoperative laboratories including a complete blood count nightly starting on Post-operative Day 0 (POD0) until drains were removed. Drain outputs were recorded three times a day (per 8 hour shift). Drains were discontinued if output was <30 mL per shift or by the end of postoperative day 2. The recorded characteristics included the following: age, sex, American Society of Anesthesiologists’ physical status classification (ASA), weight, height, body mass index (BMI), and pertinent preoperative laboratory values (prothrombin time/ international normalized ratio, creatinine, platelet count, hematocrit, and hemoglobin).

Outcome Measures

The primary outcome was reduction of Hgb, which was the difference between preoperative and the lowest postoperative Hgb values during the inpatient admission. Secondary outcomes included calculated blood loss, reduction of hematocrit, drain output, rate of postoperative transfusion, thromboembolic event, infections and length of hospital stay. In addition, intraoperative measures such as case length, calculated blood loss, intravenous fluid received and number receiving intraoperative cell saver and blood transfusion were investigated. Blood loss was calculated as a function of patient characteristics including sex, weight, and height as well as preoperative and postoperative hemoglobin balance.12,13

Sample Size and Statistical Analysis

A pretest power analysis determined that 30 patients in each group were needed to show a 1.0 g/dL difference in hemoglobin drop, assuming an equivalence margin of ± 1.0 g/dL, 5% alpha error, and 80% power. The primary outcome of Hgb drop and the secondary outcome of calculated blood loss were tested for equivalence using a two one-sided test. The remaining secondary outcomes and covariates were compared using traditional t tests. Ordinal scale outcome variables were tested using nonparametric methods such as χ2 or Fisher exact test. A P value of < 0.05 suggests statistical significance. All data were analyzed by Microsoft Excel 2011 (Microsoft, Seattle, WA) and XLSTAT 2017: Data Analysis and Statistical Solution for Microsoft Excel (Addinsoft, Paris, France).

RESULTS

During the period of study enrollment from February 2017 and March 2020, 300 patients were scheduled for elective thoracolumbar fusion surgery, which included spinal stenosis, spondylolysis, spondylolisthesis, degenerative disc disease, adult spinal deformity, proximal junctional kyphosis, adjacent level disease, and pseu-doarthrosis. 39 patients were ineligible due to meeting exclusion criteria, refusal to participate or missing pertinent laboratory values (Figure 1). Among the 261 enrolled study participants who underwent randomization, 137 had IV TXA and 124 had PO TXA. No patient was lost or excluded during the follow-up period. Two orthopedic spine surgeons performed the operations: G.G. performed majority of the surgeries (97% of total, 98% and 95% for IV and PO, respectively), and S.B. performed the rest. Standard posterior approach was utilized in all cases except for one DLIF that was performed in the PO group.

F1
Figure 1:
CONSORT (Consolidated Standards of Reporting Trials) flow diagram of the study. TXA indicates tranexamic acid.

The average age of enrolled patients was 62 ± 13 years (mean ± SD). There were 141 females and 120 males. There were no statistical differences in the patient characteristics and preoperative measurements pertaining to age, sex, height, ASA, estimated blood volume, preoperative anticoagulant use, and pertinent preoperative laboratory values including hemoglobin, hematocrit, INR, and creatinine (Table 1). Weight was higher in PO TXA group, so BMI was higher in the PO group (32.0 ± 5.8 kg/m2) than in the IV TXA group (29.5 ± 5.9 kg/m2) (P= 0.0006). Furthermore, preoperative platelet count was higher in PO group (246 ± 70 × 103 cells/mm3) than in IV group (224 ± 77 × 103 cells/mm3) (P= 0.02). Revision cases comprised of 67% of the total sample (66% and 67% for IV and PO, respectively). A single-level pedicle subtraction osteotomy was performed for 18 patients (12 and six for IV and PO, respectively) by G.G. TLIF was performed for 22 patients (11 and 19 for IV and PO, respectively). Intraoperative measures between the treatment groups demonstrated no statistical difference in anesthesia time, surgery time, IVF, percentage of patients receiving cell saver, and percentage of patients receiving blood transfusion. Overall, the patients within the two treatment groups were considered similar in regard to the measured independent variables (Table 2).

TABLE 1 - Patient Characteristics
IV PO P
No. of patients 137 124
Mean age, y 64 (12) 61 (13) 0.07
Sex (M/F) 64/73 56/68 0.80
Weight, kg 85 (20) 92 (21) 0.005
Height, m 1.69 (0.11) 1.69 (0.11) 0.9
Body mass index, kg/m2 29.5 (5.9) 32.0 (5.8) 0.0006
Estimated blood volume, mL 4916 (972) 5144 (1050) 0.07
ASA 2.7 (0.5) 2.7 (0.5) 0.85
 Class I (n) 0 2
 Class II (n) 40 31
 Class III (n) 93 90
 Class IV (n) 4 1
No. of fused levels 4.0 (2.8) 3.7 (2.8) 0.38
 One to two levels 57 62
 Three to five levels 44 34
 More than five levels 36 28
No. (%) of revisions 91 (66%) 83 (67%) 0.93
No. (%) of interbody fusions 11 (8%) 19 (15%) 0.07
No. (%) of osteotomies 19 (14%) 14 (11%) 0.53
No. (%) of anticoagulant use 12 (9%) 6 (5%) 0.21
Preoperative Hgb, g/dL 13.4 (1.5) 13.6 (1.7) 0.34
Preoperative Hct (%) 40.0 (4.4) 40.5 (4.6) 0.34
Preoperative platelet count (×103 cells/mm3) 224 (77) 246 (70) 0.02
Preoperative INR 1.07 (0.12) 1.05 (0.16) 0.26
Preoperative creatinine, mg/dL 0.97 (0.71) 0.92 (0.74) 0.6
Mean (SD). ASA indicates American Society of Anesthesiologists; INR, international normalized ratio; IV, intravenous; PO, oral.
Statistical analysis for comparison between groups: P < 0.05, statistical significance.
χ2 test or Fisher exact test.

TABLE 2 - Intraoperative Measures
IV (N = 137) PO (N = 124) P
Anesthesia time, min 372 (104) 356 (98) 0.2
Surgical time, min 275 (101) 266 (97) 0.45
Intravenous fluid, mL 3402 (1759) 3365 (1449) 0.85
No. (%) receiving cell saver 54 (39%) 42 (34%) 0.35
No. (%) receiving blood transfusion 9 (7%) 3 (2%) 0.11
Mean (SD). IV indicates intravenous; PO, oral.
Statistical analysis for comparison between groups: P < 0.05, statistical significance.
χ2 Test or Fisher exact test.

Primary Outcome Measure

The reduction in Hgb was statistically equivalent between IV and PO TXA (Table 3). The reduction in hemoglobin for the IV TXA group was 3.56 ± 1.93 g/dL while the drop in Hgb for the PO TXA group was 3.28 ± 1.60 g/dL (P= 0.002, equivalent).

TABLE 3 - Outcome Measurements
IV (N = 137) PO (N = 124) P
Hgb drop, g/dL 3.56 (1.93) 3.28 (1.60) 0.002
Calculated blood loss, mL 1270 (677) 1219 (610) 0.001
Hct drop (%) 10.5 (5.8) 9.9 (5.4) <0.0001
Drain output, mL 676 (452) 651 (469) 0.41
No. (%) postop transfusion 26 (19%) 12 (10%) 0.03
No. (%) DVT/PE 3 (2%) 3 (2%) 0.90
No. (%) infections 8 (6%) 5 (4%) 0.50
Length of hospital stay, days 4.4 (2.8) 3.7 (2.4) 0.02
Mean (SD). DVT indicates deep vein thrombosis; IV, intravenous; PE, pulmonary embolism; PO, oral.
Statistical analysis for comparison between groups: P < 0.05, statistical significance.
Welch two one-sided test: P < 0.05 demonstrates equivalence between treatments.
χ2 Test or Fisher exact test.

Secondary Outcome Measures

The calculated blood loss between the IV and PO TXA groups was equivalent (Table 3). On average, the volume of blood loss for the IV and PO TXA groups was 1270 ± 677 mL and 1219 ± 610 mL (P= 0.001, equivalent), respectively. Hematocrit drop between IV and PO TXA groups was also equivalent (Table 3). The change in hematocrit for IV and PO groups was calculated to be 10.5% ± 5.8% and 9.9% ± 5.4% (P < 0.001, equivalent), respectively. Postoperative rate of transfusion, drain output, length of hospital stay, and complications demonstrated no statistical difference between IV and PO TXA groups (Table 3). Total drain output for IV and PO TXA groups was similar at 676 ± 452 mL and 651 ± 469 mL (P= 0.41), respectively. IV TXA group had a higher postoperative transfusion rate compared to PO TXA group (22 patients [19%] vs. 12 patients [10%]; P= 0.03). Three patients (2%) in each group experienced a deep vein thrombosis/PE (P= 0.90). Eight patients in IV TXA group (6%) and five patients in PO TXA group (4%) had surgical site infections (P = 0.50). Length of hospital stay was longer in IV group compared to PO group, 4.4 ± 2.8 days and 3.7 ± 2.4 days (P= 0.02), respectively.

Subgroup Analysis

The total sample was subcategorized by the number of vertebral levels arthrodesed (one to two levels levels, three to five levels, more than five levels). The sample size was big enough in each subgroup to allow independent statistical analysis (Table 4A–C). The one- to two-level fusion group and the more than five-level fusion group demonstrated that IV TXA was equivalent to PO TXA (Hgb drop: 2.34 ± 1.35 vs. 2.73 ± 1.32 g/dL [P < 0.01] and 4.97 ± 1.65 vs. 4.79 ± 1.65 g/dL [P= 0.03], respectively; equivalence). However, the three- to five-level fusion group showed that IV TXA had higher hemoglobin drop compared to PO (4.22 ± 1.54 vs. 3.04 ± 1.22 g/dL; P < 0.001). The secondary outcomes of calculated blood loss and change in hematocrit showed similar trends (Table 4A–C). Additionally, BMI, revision, interbody fusion, osteotomy, anticoagulant use, anesthesia and surgical time can all affect blood loss (Table 5A–C). In the one- to two-level fusion group and the more tan five-level fusion group, BMI was higher in PO than IV TXA (32.6 ± 5.6 vs. 29.4 ± 6.3 kg/m2 [P < 0.005] and 32.1 ± 6.9 vs. 28.2 ± 5.2 kg/m2 [P= 0.01], respectively). In the three- to five-level fusion group, number (%) of anticoagulant use was higher in IV versus PO TXA (14% vs. 0; P= 0.03) and anesthesia time was higher in IV versus PO TXA (372 ± 71 vs. 340 ± 60 minutes; P= 0.04).

TABLE 4 - Subgroup Outcome Measurements
4A. Outcome Measurements for One to Two Fused Levels
IV (N = 57) PO (N = 62) P ∗∗
Hgb drop, g/dL 2.34 (1.35) 2.73 (1.32) <0.01
Calculated blood loss, mL 850 (436) 1020 (514) 0.01
Hct drop (%) 6.6 (3.6) 8.1 (4.0) 0.02
4B. Outcome Measurements for Three to Five Fused Levels
IV (N = 44) PO (N = 34) P
Hgb drop, g/dL 4.22 (1.54) 3.04 (1.22) <0.001
Calculated blood loss, mL 1559 (582) 1115 (444) <0.001
Hct drop (%) 12.7 (4.6) 9.2 (3.7) <0.001
4C. Outcome Measurements for More Than Five Fused Levels
IV (N = 36) PO (N = 28) P ∗∗
Hgb drop, g/dL 4.97 (1.65) 4.79 (1.65) 0.03
Calculated blood loss, mL 1706 (612) 1783 (651) 0.03
Hct drop (%) 14.7 (5.0) 14.7 (7.0) 0.04
Mean (SD). IV indicates intravenous; PO, oral.
∗∗Welch two one-sided test: P < 0.05 demonstrates equivalence between treatments.
Statistical analysis for comparison between groups: P < 0.05, statistical significance.

TABLE 5 - Subgroup Baseline Characteristics
5A. Baseline Characteristics for One to Two Fused Levels
IV (N = 57) PO (N = 62) P
BMI, kg/m2 29.4 (6.3) 32.6 (5.6) <0.005
No. (%) of revisions 38 (67%) 36 (58%) 0.33a
No. (%) of interbody fusions 5 (9%) 11 (18%) 0.15a
No. (%) of osteotomies 0 0 N/A
No. (%) of anticoagulant use 4 (7%) 4 (6%) 0.9a
Anesthesia time, min 302 (55) 303 (49) 0.97
Surgical time, min 206 (50) 211 (47) 0.58
5B. Baseline Characteristics for Three to Five Fused Levels
IV (N = 44) PO (N = 34) P
BMI, kg/m2 30.6 (6.0) 30.9 (5.4) 0.84
No. (%) of revisions 27 (61%) 22 (65%) 0.76a
No. (%) of interbody fusions 3 (7%) 6 (18%) 0.14a
No. (%) of osteotomies 1 (2%) 2 (6%) 0.41a
No. (%) of anticoagulant use 6 (14%) 0 0.03a
Anesthesia time, min 372 (71) 340 (60) 0.04
Surgical time, min 272 (67) 253 (60) 0.18
5C. Baseline Characteristics for More Than Five Fused Levels
IV (N = 36) PO (N = 28) P
BMI, kg/m2 28.2 (5.2) 32.1 (6.9) 0.01
No. (%) of revisions 26 (72%) 25 (89%) 0.09a
No. (%) of interbody fusions 3 (8%) 2 (7%) 0.86a
No. (%) of osteotomies 18 (50%) 12 (43%) 0.57a
No. (%) of anticoagulant use 2 (6%) 2 (7%) 0.79a
Anesthesia time, min 482 (102) 493 (88) 0.67
Surgical time, min 387 (99) 402 (86) 0.52
Mean (SD). BMI indicates body mass index; IV, intravenous; PO, oral.
Statistical analysis for comparison between groups: P < 0.05, statistical significance.
aχ2 Test or Fisher exact test.

DISCUSSION

Spinal fusion surgery has been associated with significant perioperative blood loss, which leads to elevated transfusion rates and serious complications.14,15 Greater health care costs result from cost of blood products, increased hospital stay and complications.4 Tranexamic acid has been studied and utilized as a hemostatic agent since the 1960 s, and its use in spine surgery ranges from pediatric to adults, from routine degenerative cases to complex deformity cases. A plethora of studies has demonstrated TXA's effects in reducing blood loss and transfusion rates.5 Although most studies focused on the intravenous and topical forms of TXA,8,16 this is one of the first studies in the spine literature that compares IV versus PO formulations of TXA. We found no difference in the efficacy between the two routes of administration in both the primary outcome of hemoglobin drop and secondary outcomes. PO TXA appears to be as safe as IV TXA in regards to postoperative thromboembolic events and infections. Furthermore, PO TXA was associated with lower transfusion rate compared to IV group.

Compared to total joints literature, where all formulations of TXA have been shown to be effective,9,10,17 the spine literature has been limited to the IV route of administration. Spine surgeons preferred to use IV TXA, especially in larger cases due to larger exposures and more extensive dissection. In both the small (one to two level) and large (more than five level) fusion subgroups, BMI was higher in the PO TXA treatment group. Previous studies indicated that higher BMI was associated with increased bloodloss,18,19 but our subanalysisshowed that PO and IV TXA had similar blood loss for the small and large fusion subgroups, which provided further evidence that PO TXA is effective at reducing blood loss. However, PO TXA was more effective than IV TXA at reducing blood loss in three- to five-level fusions. In this subgroup, IV TXA group had more anticoagulant use compared to PO TXA group (14% vs. 0%, P= 0.03). Also, surgery was longer in IV group compared to PO group (372 ± 71 vs. 340 ± 60 minutes, P= 0.04). Similarly, Peters et al20 showed that IV TXA became increasingly effective with higher fused levels and longer surgical time. Although anticoagulants were stopped according to protocol in all of our patients, their systemic clearance cannot be accurately predicted in the patients.

Given the ever-changing landscape of medical reimbursement, cost effective health care practice is beneficial to patients, providers, the health care system and society at large. The use of TXA in orthopedic surgery has become a routine practice because it has been shown to be safe, clinically beneficial, and cost-effective.21 TXA offers not only direct savings from transfusion costs22 but also inpatient hospital cost that is related to length of stay.23 Given that PO TXA is cheaper and easier to administer than IV TXA, switching to PO can lead to greater cost savings. In our institution, the oral TXA dosage cost $14 compared with $53 for the generic IV formulation alone (not including the cost of pharmacy preparation). Reducing length of stay with PO TXA can generate even greater cost savings. As the American population continues to age, the number of spinal fusion surgeries performed in the United States will likely continue to increase from the current annual rate of about 500,000. Consequently, transitioning to PO TXA has potential to yield cost savings of at least 20 million dollars per year for the health care system.

Our study is not without potential limitations. First, the study population contains heterogeneity such as varying patient diagnosis and surgical technique/approach. However, heterogeneity was minimized through substratification of the number of fusion levels into three categories. Posterior approach was mostly used (100% and 99% for IV and PO, respectively), and a single surgeon performed most of the cases (98% and 95% for IV and PO, respectively). Second, blood loss calculation was based on the lowest postoperative hemoglobin value, which may be inaccurate due to hemodilution if the patient was discharged before postoperative day five.17 However, change in hematocrit was similar for IV and PO groups. Despite these limitations, the validity of our results should be maintained, as the same methodology was applied to both treatment arms.

The third potential limitation is that we did not include a placebo group and assumed that PO TXA was superior to placebo based on current literature. We believe that using the standardIV formulationasacontrol insteadofa pure placebo was more clinically useful. Additionally, because perioperative bleeding during spine surgery is multifactorial and can be more significant than TKA or THA,24 we would put patients at increased risk for blood loss if we did not give TXA. Additionally, administration of TXA in spine surgery is already emerging as the standard of care, and this study is an attempt to optimize current standards. Lastly, the large percentage (67%) of revision cases in our cohort is a limitation as the results are not directly generalizable to primary cases. Strengthsofthis study include thatitwascompletedata single center and by a single surgeon predominantly.

CONCLUSION

In conclusion, in the setting of spine thoracolumbar fusions, oral TXA produced an equivalent reduction in hemoglobin and blood loss compared to its intravenous counterpart. Given the equivalent clinical outcomes, potential hospital cost savings, and the ease of drug administration, oral TXA is a superior alternative to intravenous TXA.

Key Points

  • The mean reduction of hemoglobin and calculated blood loss were equivalent between IV and PO TXA groups.
  • Postoperative complication rates were similar between the two groups, but transfusion rate was higher in IV TXA group. Length of stay was longer in IV TXA group.
  • PO TXA is a superior alternative to IV TXA in patients undergoing elective posterior thoracolumbar fusion.

References

1. Zollo RA, Eaton MP, Karcz M, et al. Blood transfusion in the perioperative period. Best Pract Res Clin Anaesthesiol 2012; 26:475–484.
2. Marik PE, Corwin HL. Efficacy of red blood cell transfusion in the critically ill: a systematic review of the literature. Crit Care Med 2008; 36:2667–2674.
3. Yerneni K, Burke JF, Tuchman A, et al. Topical tranexamic acid in spinal surgery: a systematic review and meta-analysis. J Clin Neurosci 2019; 61:114–119.
4. Hofmann A, Ozawa S, Farrugia A, et al. Economic considerations on transfusion medicine and patient blood management. Best Pract Res Clin Anaesthesiol 2013; 27:59–68.
5. Cheriyan T, Maier SP, Bianco IIK, et al. Efficacy of tranexamic acid on surgical bleeding in spine surgery: a meta-analysis. Spine J 2015; 15:752–761.
6. Yang B, Li H, Wang D, et al. Systematic review and meta-analysis of perioperative intravenous tranexamic acid use in spinal surgery. PLoS One 2013; 8:e55436.
7. Li Z-J, Fu X, Xing D, et al. Is tranexamic acid effective and safe in spinal surgery? A meta-analysis of randomized controlled trials. Eur Spine J 2013; 22:1950–1957.
8. Winter SF, Santaguida C, Wong J, et al. Systemic and topical use of tranexamic acid in spinal surgery: a systematic review. Global Spine J 2016; 6:284–295.
9. Fillingham YA, Kayupov E, Plummer DR, et al. A randomized controlled trial of oral and intravenous tranexamic acid in total knee arthroplasty: the same efficacy at lower cost? J Arthroplasty 2016; 31: (9 suppl): 26–30.
10. Kayupov E, Fillingham Y, Okroj K. Oral and intravenous tranexamic acid are equivalent at reducing blood loss following total hip arthroplasty. J Bone Joint Surg Am 2017; 99:373–378.
11. Yu CC, Kadri O, Kadado A, et al. Intravenous and oral tranexamic acid are equivalent at reducing blood loss in thoracolumbar spinal fusion: a prospective randomized trial. Spine (Phila Pa 1976) 2019; 44:755–761.
12. Mercuriali F, Inghilleri G. Proposal of an algorithm to help the choice of the best transfusion strategy. Curr Med Res Opin 1996; 13:465–478.
13. Nadler SB, Hidalgo JH, Bloch T. Prediction of blood volume in normal human adults. Surgery 1962; 51:224–232.
14. Neilipovitz DT. Tranexamic acid for major spinal surgery. Eur Spine J 2004; 13: (suppl 1): S62–S65.
15. Elgafy H, Bransford RJ, McGuire RA, et al. Blood loss in major spine surgery: are there effective measures to decrease massive hemorrhage in major spine fusion surgery? Spine (Phila Pa 1976) 2010; 35:S47–56.
16. Xiong Z, Liu J, Yi P, et al. Comparison of intravenous versus topical tranexamic acid in non-deformity spine surgery: a meta-analysis. Biomed Res Int 2020; 2020:12.
17. Gomez-Barrena E, Ortega-Andreu M, Padilla-Eguiluz NG, et al. Topical intra-articular compared with intravenous tranexamic acid to reduce blood loss in primary total knee replacement: a double-blind, randomized, controlled, noninferiority clinical trial. J Bone Joint Surg Am 2014; 96:1937–1944.
18. De la Garza-Ramos R, Bydon M, Abt NB, et al. The impact of obesity on short- and long-term outcomes after lumbar fusion. Spine (Phila Pa 1976) 2015; 40:56–61.
19. Jiang J, Teng Y, Fan Z, et al. Does obesity affect the surgical outcome and complication rates of spinal surgery? A meta-analysis. Clin Orthop Relat Res 2014; 472:968–975.
20. Peters A, Verma K, Slobodyanyuk K, et al. Antifibrinolytics reduce blood loss in adult spinal deformity surgery: a prospective, randomized controlled trial. Spine (Phila Pa 1976) 2015; 40:E443–E449.
21. Danninger T, Memtsoudis SG. Tranexamic acid and orthopedic surgery-the search for the holy grail of blood conservation. Ann Transl Med 2015; 3:77.
22. Moskal JT, Harris RN, Capps SG. Transfusion cost savings with tranexamic acid in primary total knee arthroplasty from 2009 to 2012. J Arthroplasty 2015; 30:365–368.
23. Gillette BP, Maradit Kremers H, Duncan CM, et al. Economic impact of tranexamic acid in healthy patients undergoing primary total hip and knee arthroplasty. J Arthroplasty 2013; 28: (8 suppl): 137–139.
24. Dekutoski MB. Blood loss and transfusion management in spinal surgery. Orthopedics 1999; 22: (1 suppl): s155–s157.
Keywords:

allogeneic blood transfusion; blood loss; healthcare cost efficiency; spinal fusion; tranexamic acid

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.