Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Apr 1.
Published in final edited form as: J Pediatr Hematol Oncol. 2022 Apr 1;44(3):e701–e706. doi: 10.1097/MPH.0000000000002351

Blood loss and transfusion in a pediatric scoliosis surgery cohort in the antifibrinolytic era

Carolyn G Ahlers 1, Matthews Lan 1, Jonathan G Schoenecker 2, Alexandra J Borst 3
PMCID: PMC8957516  NIHMSID: NIHMS1741824  PMID: 34654764

Abstract

Children and adolescents undergoing posterior spinal fusion for scoliosis experience high rates of bleeding and blood product transfusion. Antifibrinolytic therapy is one key strategy to decrease blood loss and transfusion in pediatric scoliosis surgery. Here we review 172 pediatric scoliosis patients (birth to 21 years) who underwent posterior spinal fusion at our institution from 2017 to 2018. We reported rates of blood loss and transfusion, compared patients receiving tranexamic acid (TXA) to a ε-aminocaproic acid (EACA), and evaluated antifibrinolytic agent and laboratory parameters as predictors of blood loss and transfusion. Intraoperatively, 62% received TXA and 38% received EACA. Overall, blood loss (mean intraoperative estimated blood loss = 14.9 ± 9.7 mL/kg, 22% with clinically significant blood loss (> 20mL/kg), and mean calculated Hgb mass loss = 175.9 ± 70.1 g) and transfusion rates (15% with intraoperative allogeneic red blood cell [RBC] transfusion and mean intraoperative allogeneic RBC transfusion volume = 12.5 ± 7.1 mL/kg) were similar to previous cohorts studying intraoperative antifibrinolytics. There was no difference in intraoperative estimated blood loss, clinically significant blood loss, calculated hemoglobin mass loss, or transfusion rates between the antifibrinolytic groups. Antifibrinolytic choice was not predictive of blood loss or transfusion. Routine hematologic laboratory parameters and antifibrinolytic choice were insufficient to predict blood loss or other outcomes. Future prospective laboratory-based studies may provide a more comprehensive model of surgical-induced coagulopathy in scoliosis surgery and provide a better tool for predicting blood loss and improving outcomes.

Keywords: Antifibrinolytic, transfusion, scoliosis, bleeding, pediatric

Introduction

Children and adolescents undergoing posterior spinal fusion surgery for scoliosis experience high rates of bleeding,[1,2] necessitating blood product transfusion in 30–90% of patients.[1,3,4] High rates of bleeding and transfusion confer an increased risk for postoperative complications and mortality.[1,3,5] Additionally, high transfusion rates put children at-risk for potentially life-threatening reactions including alloimmunization, as well as utilize an important limited resource.[4] Age, number of spinal levels fused, surgical approach, and neuromuscular scoliosis subtype are all known risk factors for increased blood loss in scoliosis surgery.[1–3,6] However, the blood loss in scoliosis surgery has also been likened to a type of controlled traumatic coagulopathy, for which the exact mechanisms are not well understood.[1,7–9] Efforts to improve our understanding of the mechanisms of this surgical-induced coagulopathy and intervention to decrease bleeding and transfusion are likely to improve outcomes for children undergoing surgical repair of scoliosis.

A key strategy for decreasing blood loss and transfusion in scoliosis surgery is the use of antifibrinolytic agents.[2,5,10] Tranexamic acid (TXA) and ε-aminocaproic acid (EACA) are synthetic lysine analogs that bind to and prevent conversion of plasminogen to plasmin, preventing fibrin degradation. TXA and EACA are used clinically in many settings where hyperfibrinolysis leads to excessive bleeding.[11,12] Antifibrinolytics reduce intraoperative blood loss and transfusion burden in scoliosis surgery,[13] and their efficacy has now been well documented in a variety of surgical settings.[13–17] Efforts are ongoing to determine the optimal dosing regimen, duration of therapy, and relative efficacy of TXA vs EACA in pediatric scoliosis surgery.[13,17–20]

The primary goal of this single-center retrospective study was to describe blood loss and transfusion rates in a pediatric scoliosis cohort and to compare these outcomes in patients receiving TXA versus EACA. An important secondary goal was to evaluate if antifibrinolytic choice and preoperative hematologic laboratory parameters were predictive of blood loss and transfusion, as this knowledge could be useful in guiding future optimal antifibrinolytic regimens, understanding the mechanisms of surgical-induced coagulopathy, informing perioperative optimization efforts, and improving outcomes for children with scoliosis.

Materials and Methods

Data Collection

This was a single-institution retrospective cohort study of pediatric scoliosis patients who underwent posterior spinal fusion during a two-year period (2017–2018). This study was approved by the Institutional Review Board of Vanderbilt University Medical Center (IRB #170551). Only patients who received intraoperative antifibrinolytic therapy as per institutional standard of care were included. Patients with a diagnosed inherited bleeding disorders or history of venous thromboembolism were excluded. In this consecutive case series, only one patient was excluded who had an underlying diagnosis of congenital factor VII deficiency. Demographic and clinical information were recorded from the electronic medical record.

Neuromuscular scoliosis was defined as a non-congenital spinal deformity that occurs in patients with any type of pre-existing neuromuscular diagnosis.[21] Idiopathic scoliosis was defined as scoliosis in which the underlying etiology is unknown or not related to a specific syndrome, congenital, or neuromuscular condition.[6] Antifibrinolytic type (TXA, EACA) and dosage (bolus and continuous) were recorded. Patients receiving EACA received a bolus dose of 100mg/kg (max 5g) followed by infusion of 10–30mg/kg/hour. There was one outlier who received a bolus of 4500g and one patient who received no bolus and an infusion rate of 0.5mg/kg/hr. There was no documentation of rationale for the 4500g dose, but it was documented in both the anesthetic record and medication administration record as 4500 grams, not milligrams. At this time, dosing was up to the discretion of the anesthesia and surgical teams and was not protocolized, but 64/66 (97%) of patients received dosing within the formulary guidelines. TXA dosing was a 50mg/kg (max 2000mg) bolus with an infusion rate of 10 mg/kg/hour. There were three outliers with bolus dosing of 10mg/kg, 4000mg, and 8500mg. Of note, EACA was given at our institution for most of 2017, while in 2018, there was a shift toward using TXA, primarily due to drug availability. Perioperative laboratory variables collected included hemoglobin, platelet count, fibrinogen, activated partial thromboplastin time (aPTT), and prothrombin time (PT). Blood loss measurements included intraoperative estimated blood loss per kilogram (kg) (mL/kg) which was estimated by the clinical anesthesia team and calculated hemoglobin (Hgb) mass loss (g), which has been found to be a more accurate method to estimate perioperative blood loss and is calculated as mHgbEBL = 100 × (Hgbpreop – Hgbnadir) × BV (mHgbEBL [g] is estimated hemoglobin mass loss, Hgbpreop [g/dL] is the patient’s preoperative hemoglobin, Hgbnadir [g/dL] is the patient’s lowest postoperative hemoglobin, and BV [mL] is the patient’s estimate blood volume).[22,23] Clinically significant blood loss was defined as > 20mL/kg, as a transfusion volume of 20 mL/kg or greater is known to be associated with increased rates of total complications.[24] Transfusion volumes (mL/kg) of packed red blood cells (allogeneic RBC), fresh frozen plasma (FFP), cryoprecipitate, and platelets in the intraoperative and postoperative periods were recorded. At our institution, standard anesthesia protocol calls for autologous transfusion at the end of each spinal fusion case regardless of estimated intraoperative blood loss. The trigger for allogeneic RBC blood transfusion at our institution is hemoglobin < 8g/dL or hematocrit < 24% for all patients, with anesthesia team discretion. Adverse events within 30 days of surgery were recorded. This included wound dehiscence, wound infection, readmission for any issue, thrombotic events, and bleeding events. Bleeding events were defined as any intraspinal bleeding, bleeding into any other anatomic cavity, bleeding requiring re-operation, clinically significant wound hematoma, or gastrointestinal bleed. Thrombotic events were defined as any arterial or venous thrombus documented by imaging study or cerebrovascular accident. All blood loss and transfusion data were recorded as mL/kg.

Statistical Analysis

Data were analyzed using Statistical Package for Social Sciences (SPSS) Version 26.0 (Armonk, NY. IBM Corp.). Continuous variables were presented as mean and standard deviation (SD) or median and interquartile range (IQR) for variables with non-normal distribution (such as number of spinal levels fused). Categorical variables were presented as frequency and percentage. For univariate analysis, chi-square tests were used to evaluate differences in categorical data among patients who received intraoperative TXA versus EACA. Continuous variables were compared using Student’s t-tests. Multivariate linear regression analyses were used to evaluate antifibrinolytic choice and preoperative hematologic laboratory parameters as predictors of intraoperative estimated blood loss (ml/kg) and calculated Hgb mass loss (g). Univariate regression analysis was used for the outcomes of intraoperative allogeneic RBC transfusion (n=26) and postoperative allogeneic RBC transfusion (n=21) due to low sample size limiting ability to control for multiple covariates. Clinically significant blood loss was a binary outcome with patients being either at/above or below the threshold of 20 mL/kg reported blood loss. Logistic regression analysis was used for this outcome. Independent variables included antifibrinolytic agent, scoliosis diagnosis (idiopathic vs. neuromuscular), preoperative hemoglobin, preoperative platelet count, and preoperative fibrinogen. An alpha level was set a priori at 0.05. In the multivariate analyses, we controlled for scoliosis diagnosis, age, weight, height, sex, number of spinal levels fused, and surgery time, as these factors are all known factors that influence blood loss.[3,6] Bolus and continuous dosages of each antifibrinolytic medication were also controlled for in the multivariate analyses investigating the predictive role of TXA versus EACA in affecting blood loss and transfusions. Odds ratios and 95% confidence intervals are reported for the univariate logistic regression analysis of clinically significant blood loss. The results of all linear regression analyses are reported as positive or negative predictors with p-values (if statistically significant).

Results

Cohort Demographics

A total of 172 pediatric scoliosis patients were included in this retrospective cohort study: 59% (n=102) had idiopathic scoliosis, 25% (n=43) had neuromuscular scoliosis (NMS), and 16% (n=27) were classified as “other”. In the “other” diagnosis group (Supplemental Table 1), 44% (n=12) of patients had congenital scoliosis, 19% (n=5) of patients had scoliosis due to osteogenesis imperfecta, and 19% (n=5) had Scheuermann’s kyphosis. Of the 172 patients in this study, 62% (n=106) received TXA and 38% (n=66) received EACA. There were no significant differences between the groups in regard to sex, race, diagnosis, age, weight, height, number of spinal levels fused, or surgery time. (Table 1). Importantly, there was no significant difference in the proportions of scoliosis diagnosis between the two antifibrinolytic groups (p=0.789).

Table 1:

Cohort demographics, surgical information, and blood loss measurements by antifibrinolytic group

Characteristic Total (n=172) EACA (n=66) TXA (n=106) p-value
Female, n (%) 109 (63%) 40 (61%) 69 (65%) 0.552
Race/Ethnicity, n (%)
 White 140 (81%) 55 (83%) 85 (80%) 0.922
 Black or African American 27 (16%) 9 (14%) 18 (17%)
 Asian 3 (2%) 1 (2%) 2 (2%)
 Other 2 (1%) 1 (2%) 1 (1%)
Diagnosis, n (%)
 Idiopathic scoliosis 102 (59%) 37 (56%) 65 (61%) 0.789
 Neuromuscular scoliosis 43 (25%) 18 (27%) 25 (24%)
 Other 27 (16%) 11 (17%) 16 (15%)
Age (years), mean (SD) 13.5 (3.1) 13.5 (2.7) 13.5 (3.3) 0.994
Weight at time of surgery (kg), mean (SD) 50.3 (20.7) 49.4 (19.7) 50.9 (21.4) 0.653
Height at time of surgery (cm), mean (SD) 149.9 (22.5) 150.6 (21.5) 149.5 (23.1) 0.753
Number of Spinal Levels Fused, median (IQR) 11 (9–13) 12 (9–14.25) 11 (9–13) 0.144
Duration of surgery (min), mean (SD) 389.1 (83.1) 392.2 (71.7) 387.2 (89.7) 0.703
Total intraoperative blood loss (mL/kg), mean (SD) 14.9 (9.7) 13.2 (6.8) 14.9 (9.7) 0.158
Estimated hemoglobin mass loss (g), mean (SD); n=141 total, 84 idiopathic, 35 NMS, 21 other 175.9 (70.1) 187.4 (55.4) 171.5 (74.7) 0.229
Clinically significant blood loss (> 20ml/kg), n (%) 38 (22%) 13 (20%) 25 (24%) 0.550
Open in a new tab
†

Statistically significant at p < 0.05

EACA = ε-aminocaproic acid, TXA = tranexamic acid, SD = standard deviation, IQR = interquartile range, NMS = neuromuscular scoliosis

Blood Loss

Mean intraoperative estimated blood loss for all patients was 14.9 ± 9.7 mL/kg, and clinically significant blood loss (> 20mL/kg) occurred in 22% of patients. The mean calculated Hgb mass loss[22] for all patients was 175.9 ± 70.1 g. There was no difference between the antifibrinolytic groups for intraoperative estimated blood loss, calculated Hgb mass loss, or clinically significant blood loss (Table 1).

Transfusion

Nearly all patients (91%, n=157) received intraoperative autologous transfusion via cell saver. This was consistent with anesthesia protocol to return autologous blood volume to patients at the end of the case regardless of blood loss. Intraoperative allogeneic RBC transfusion was only given to 15% (n=26) patients and only 1% of patients received intraoperative FFP (n=2) and platelet transfusions (n=1). Mean intraoperative autologous transfusion volume per kg was 4.0 ± 2.7 mL/kg and mean intraoperative allogeneic RBC transfusion volume per kg was 12.5 ± 7.1 mL/kg. Postoperatively, 12% (n=21) of patients received allogeneic RBC transfusion, while only 1% of patients received postoperative platelet (n=1) and FFP transfusions (n=2). The majority (n=21/23, 91%) of patients who received any postoperative transfusion (pRBC, FFP or platelets) had previously received intraoperative transfusion. One patient who required postoperative FFP and platelets had received both pRBCs and cell saver intraoperatively. The patient who required only FFP postoperatively had only received cell saver intraoperatively. Mean postoperative allogeneic RBC transfusion volume per kg was 11.5 ± 5.4 mL/kg. There were no differences between the TXA and EACA antifibrinolytic groups in intraoperative or postoperative transfusions rates or volume per kg (Table 2).

Table 2:

Transfusion data and perioperative hematologic parameters by antifibrinolytic group

Intraoperative transfusions Total (n=172) EACA (n=66) TXA (n=106) p-value
Patients with allogeneic pRBC transfusion, n (%) 26 (15%) 9 (14%) 17 (16%) 0.669
Total allogeneic pRBCs transfused (mL/kg), mean (SD) 12.5 (7.1) 11.2 (4.4) 13.2 (8.1) 0.501
Patients with autologous blood transfusion, n (%) 157 (91%) 59 (89%) 98 (92%) 0.489
Total autologous blood transfused (mL/kg), mean (SD) 4.0 (2.7) 4.0 (2.5) 4.0 (2.8) 0.997
Postoperative transfusions Total (n=172) EACA (n=66) TXA (n=106) p-value
Patients with allogeneic pRBC transfusion, n (%) 21 (12%) 8 (12%) 13 (12%) 0.978
Total allogeneic pRBCs transfused (mL/kg), mean (SD) 11.5 (5.4) 11.1 (5.3) 11.7 (5.7) 0.806
Preoperative Lab Values Total (n=172) EACA (n=66) TXA (n=106) p-value
Pre-op hemoglobin (g/dL), mean (SD), n=168 13.7 (1.3) 13.7 (1.2) 13.6 (1.3) 0.797
Pre-op platelet count (x103/uL), mean (SD), n=167 289.2 (66.5) 287.7 (59.0) 290.1 (71.0) 0.824
Pre-op fibrinogen (mg/dL), mean (SD), p=151 311.6 (79.6) 309.4 (65.8) 313.3 (89.2) 0.771
Pre-op aPTT (s), mean (SD), p=155 30.5 (3.0) 31.1 (3.4) † 30.1 (2.6) † 0.046 †
Pre-op PT (s), mean (SD), n=164 13.8 (1.5) 13.4 (1.7) † 14.0 (1.4) † 0.019 †
Postoperative Lab Values Total (n=172) EACA (n=66) TXA (n=106) p-value
Post-op minimum hemoglobin 0–72 hrs (g/dL), mean (SD), n=144 8.3 (1.5) 8.0 (1.6) 8.4 (1.4) 0.156
Post-op minimum platelet 0–72 hrs (x103/uL), mean (SD), n=143 149.6 (53.6) 146.3 (42.1) 150.8 (57.5) 0.662
Post-op minimum fibrinogen 0–72 hrs (mg/dL), mean (SD), n=35 222.0 (79.1) 199.9 (43.2) 232.1 (90.0) 0.162
Post-op maximum aPTT 0–72 hrs (s), mean (SD), n=6 25.8 (6.4) - (-) 25.8 (6.4) - (-)
Post-op maximum PT 0–72 hrs (s), mean (SD), n=44 14.9 (1.3) 14.5 (1.1) 15.1 (1.4) 0.203
Open in a new tab
†

Statistically significant at p < 0.05

EACA = ε-aminocaproic acid, TXA = tranexamic acid, SD = standard deviation, pRBC = packed red blood cells, Pre-op = preoperative, Post-op = postoperative, aPTT = activated partial thromboplastin time, PT = prothrombin time

Adverse Events

Overall, 17% (n=30) of patients experienced a postoperative adverse event (Table 3). This included seven wound healing issues (5%), four wound infections (2%), four readmissions (2%), five bleeding events (3%), and 14 other complications (8%). Bleeding events included operative site hematoma, bleeding at tracheostomy site, and three children with gastrointestinal bleeding. There were no thrombotic events. One child with a baseline history of epilepsy experienced increased seizure frequency. This patient received TXA, which has been associated with increased seizure risk compared to EACA,[25,26] but this was after institutional transition to TXA in 2018 and EACA was not available. Other complications are detailed in Table 3. More patients in the EACA group experienced wound infection (four patients (6%) vs. zero patients (0%), p =0.010). There were no other significant differences in adverse events between the antifibrinolytic groups.

Table 3:

Adverse events by antifibrinolytic group

Adverse Event Total (n=172) EACA (n=66) TXA (n=106) p-value
Total patients with adverse events, n (%) 30 (17%) 9 (14%) 21 (20%) 0.299
Patients with wound healing issue or dehiscence events, n (%) 7 (5%) 1 (2%) 6 (6%) 0.181
Patients with wound infection, n (%) 4 (2%) 4 (6%) † 0 (0%) † 0.010 †
Patients with readmission for any issue, n (%) 4 (2%) 1 (2%) 3 (3%) 0.310
Patients with bleeding events, n (%) 5 (3%) 0 (0%) 5 (5%) 0.073
Patients with other complications, n (%) 14 (8%) 5 (8%) 9 (8%) 0.831
 Aborted surgery 1 (1%) 0 (0%) 1 (1%) 0.429
 Increased frequency of seizures 1 (1%) 0 (0%) 1 (1%) 0.429
 Respiratory Distress/Failure 7 (4%) 1 (2%) 6 (6%) 0.181
 Ileus/Urinary Retention 3 (2%) 2 (3%) 1 (1%) 0.309
 Emesis w/ fever 1 (1%) 1 (2%) 0 (0%) 0.204
 Cervicothoracic kyphosis 1 (1%) 0 (0%) 1 (1%) 0.429
 UTI 1 (1%) 1 (2%) 0 (0%) 0.204
 Urticaria 1 (1%) 0 (0%) 1 (1%) 0.429
Open in a new tab
†

Statistically significant at p < 0.05

EACA = ε-aminocaproic acid, TXA = tranexamic acid, SD = standard deviation, UTI = urinary tract infection

Hematologic Laboratory Parameters

Perioperative hematologic parameters, including hemoglobin, platelet count, fibrinogen, activated partial thromboplastin time (aPTT), and prothrombin time (PT) were collected for the cohort and summarized in Table 2. Assessment of postoperative trends was limited by fibrinogen (n=25), PT (n=43), and aPTT (n=6) being collected in a much smaller number of patients, but in general, patients did develop anemia and very mild thrombocytopenia post-operatively. Minimum hemoglobin levels in the 8–9g/dL range postoperatively can be considered to be clinically significant mild to moderate anemia. No patients developed thrombocytopenia below a platelet count of 100 ×103/uL. Preoperative aPTT and PT were significantly different between antifibrinolytic groups (Table 2), although this was unlikely to be clinically relevant. Likewise, although fibrinogen activity decreased postoperatively, there was no clinically significant drop with no patients dropping below 100mg/dL. No other laboratory parameters in the preoperative or postoperative period were different across the antifibrinolytic groups. On secondary analysis, we did find that NMS patients had significantly higher preoperative hemoglobin (14.1±1.4g/dL vs. 13.6±1.1g/dL, p=0.014) and fibrinogen (339±100mg/dL vs. 296±69mg/dL, p=0.009) compared to idiopathic scoliosis patients.

Predictors of Blood Loss and Transfusions

In order to identify if antifibrinolytic choice and preoperative hematologic laboratory parameters could predict blood loss and transfusion outcomes in pediatric scoliosis surgery, we used regression analysis to identify predictors of our blood loss outcomes. We found that a higher preoperative hemoglobin predicted increased intraoperative estimated blood loss (p=0.012) and calculated Hgb mass loss (p<0.001) and a higher preoperative fibrinogen predicted decreased intraoperative estimated blood loss (p=0.042). Univariate analysis of preoperative fibrinogen and association with intraoperative estimated blood loss was significant as demonstrated by the linear model [Intraoperative estimated blood loss (mL) = 19.87–0.016*preop Fibrinogen, p=0.042 for preop Fibrinogen]. However, on multivariate analysis, when correcting for several factors known to be associated with blood loss in scoliosis surgery,[2,7,27,28] this was no longer significant [Intraoperative estimated blood loss (mL) = 12.073+0.053*Age-0.232*Number of Spinal levels fused+2.043*sex-0.136*Weight at time of surgery-0.024*Height at time of surgery+0.043*Surgery time-0.009*preop Fibrinogen, p=0.17 for preop Fibrinogen]. None of the variables were significant predictors of clinically significant blood loss nor intraoperative or postoperative allogenic transfusion. Antifibrinolytic choice, preoperative platelet count, and preoperative PT and aPTT were not significant predictors of blood loss or transfusion in any analyses.

Discussion

In this single center retrospective cohort, pediatric patients undergoing posterior spinal fusion for scoliosis had similar rates of blood loss and transfusion to what has been reported by other groups in the era of intraoperative antifibrinolytics, and lower than pre-antifibrinolytic reports.[1,2,29–32] We also found that use of antifibrinolytics was safe, with no thrombotic complications and only one case of increased seizure frequency, albeit in a child with a known history of epilepsy. There were four patients that experienced wound site infection who had received EACA, but it is unclear if this was drug related and all patients recovered well. In the era of antifibrinolytic therapy in pediatric scoliosis surgery, our results confirm others’ findings that these drugs are safe and effective at decreasing blood loss and transfusion in scoliosis surgery.

In addition to describing blood loss and transfusion in our cohort, we sought to compare the two antifibrinolytic agents. TXA is more potent than EACA and has a longer half-life.[33] TXA has been shown to reduce blood loss and transfusion in various surgical procedures, including prospectively in scoliosis repair.[10,34] Studies comparing TXA to EACA have been relatively sparse and have shown conflicting results.[34,35] While there is clear evidence in the literature that TXA and EACA reduce blood loss and transfusion requirements in pediatric patients undergoing spinal surgery, the relative efficacy of TXA versus EACA is not clear. We found no difference in intraoperative estimated blood loss, calculated Hgb mass loss, transfusion rates or volumes, adverse events, or hematologic parameters between patients that received TXA versus EACA. This is reassuring that choice of agent is unlikely to affect outcomes significantly, although the retrospective nature of our study limits a true comparison of these agents. Future prospective studies investigating optimal dosing and duration of antifibrinolytic therapy may be beneficial in maximizing efficacy for blood loss outcomes while maintaining safety. While TXA and EACA are relatively safe and inexpensive medications, there is concern about the potential for thrombosis, increasing risk of seizures, and theoretical interference with the fibrinolysis necessary for wound healing.[10,25,26,36–38] However, more recent reports have concluded a lower seizure risk than previously thought to be associated with antifibrinolytic use and no increased risk for poor wound healing or surgical site infections [39,40]. Fortunately, in our cohort, there was only one child with seizure, who had a baseline history of epilepsy. There were no thrombotic events, and wound issues were fairly limited and not necessarily related to antifibrinolytic therapy. Although not contraindicated, a careful risk:benefit analysis should be undertaken when using antifibrinolytic agents in patients with increased risk for seizures or wound healing concerns.

Coagulopathy in scoliosis surgery is multifactorial, stemming from significant effects on hemostasis due to the large and highly vascular surgical bed, dilutional coagulopathy, dysregulation of normal counterbalancing mechanisms, hyperfibrinolysis, and a systemic inflammatory response.[7,21,36,41] However, the specific mechanisms underlying coagulopathy in scoliosis surgery are not well defined. The complex interplay between the coagulation and inflammation systems may contribute to adverse outcomes in this population, particularly in patients with NMS, where differences in coagulation systemic activation have been proposed to contribute to adverse outcomes.[21,42] Conventional pro-coagulant based assays (PT and aPTT) provide limited information about the in vivo mechanisms underlying this coagulopathy.[42,43] Additionally, the mechanisms by which antifibrinolytic agents may attenuate surgically induced hyperfibrinolysis, inflammation, and coagulopathy have not been formally investigated.

We compared perioperative laboratory parameters between the antifibrinolytic groups and found a statistically significant, but likely clinical meaningless difference in preoperative aPTT and PT between TXA and EACA groups. We did find higher preoperative hemoglobin and fibrinogen levels in the NMS group compared to idiopathic scoliosis group and found that higher preoperative hemoglobin and fibrinogen were predictive of some important blood loss and transfusion outcomes. However, when correcting for other known predictors of blood loss in scoliosis surgery, these results were not significant and these changes in hemoglobin and fibrinogen likely clinically irrelevant. Although some baseline differences in laboratory parameters have been identified in children with NMS compared to those with idiopathic scoliosis,[42,44] thought to be in part due to chronic hypoxia and differences in nutritional status, the specific finding of increased fibrinogen has not been described. In isolation, it is hard to use these relatively small changes in hemoglobin and fibrinogen to define surgical coagulopathy in this cohort. Small postoperative changes in fibrinogen, platelet count, and conventional coagulation assays are not likely significant in isolation but could be reflective of ongoing hemostatic disturbance. Further investigation of alternative coagulation and inflammatory assays, including global assays of hemostasis, may help better define the mechanisms of coagulopathy in this population.

This study was primarily limited by its retrospective nature. A direct comparison of TXA and EACA would require a future double-blinded and randomized prospective study. Given the current availability of both drugs, known differences in drug costs, and potential differences in side effect profile, this may be useful to determining both maximal efficacy and safety. Similarly, optimal dosing and duration of antifibrinolytic therapy would be best evaluated in a prospective trial. Additionally, the amount of intravenous fluid administration during surgery was not recorded in our study. While we do not believe patients were significantly hyper- or hypovolemic, it should be noted that the Hgb mass loss equation assumes euvolemia and that dilutional effects based on volume status could alter our studied hematologic outcomes, in particular the Hgb mass loss parameter, which could be dramatically affected by excessive fluid administration. There was also potential variation between specific surgeons in terms of surgical approach or time of surgery that was not accounted for. Another limitation is the inherent change in perioperative care over the timeframe of analysis. This is notable, as the type of antifibrinolytic administered was dependent on a change in drug availability at our institution over the timeframe and therefore not randomized.

Conclusion

In this study, we found no difference in blood loss, transfusion rates, or adverse outcomes for pediatric scoliosis surgery patients receiving TXA versus EACA during posterior spinal fusion. While preoperative hemoglobin and fibrinogen values predicted estimated intraoperative blood loss and Hgb mass loss in this cohort, no hematologic laboratory parameters predicted clinically significant blood loss or perioperative allogenic transfusion. Routine laboratory studies were insufficient to describe coagulopathy in this population. Further research to better understand the mechanisms involved in surgical-induced coagulopathy as well as to optimize antifibrinolytic therapy is needed to improve blood loss, transfusion, and outcomes in pediatric scoliosis surgery.

Supplementary Material

Supplemental Data File (.doc, .tif, pdf, etc.)

Acknowledgments

The authors would like to thank the patients and families who make this work possible. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Footnotes

Conflict of Interest

The authors have no conflicts of interest.

References

  • [1].Kim HJ, Park HS, Jang MJ, Koh WU, Song JG, Lee CS, et al. Predicting massive transfusion in adolescent idiopathic scoliosis patients undergoing corrective surgery. Med U S 2018;97. 10.1097/MD.0000000000010972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [2].Hassan N, Halanski M, Wincek J, Reischman D, Sanfilippo D, Rajasekaran S, et al. Blood management in pediatric spinal deformity surgery: Review of a 2-year experience. Transfusion (Paris) 2011;51:2133–41. 10.1111/j.1537-2995.2011.03175.x. [DOI] [PubMed] [Google Scholar]
  • [3].Yoshihara H, Yoneoka D. Predictors of allogeneic blood transfusion in spinal fusion for pediatric patients with idiopathic scoliosis in the United States, 2004–2009. Spine 2014;39:1860–7. 10.1097/BRS.0000000000000530. [DOI] [PubMed] [Google Scholar]
  • [4].Yoshihara H, Yoneoka D. National trends in spinal fusion for pediatric patients with idiopathic scoliosis: Demographics, blood transfusions, and in-hospital outcomes. Spine 2014;39:1144–50. 10.1097/BRS.0000000000000354. [DOI] [PubMed] [Google Scholar]
  • [5].Willner D, Spennati V, Stohl S, Tosti G, Aloisio S, Bilotta F. Spine Surgery and Blood Loss: Systematic Review of Clinical Evidence. Anesth Analg 2016;123:1307–15. 10.1213/ANE.0000000000001485. [DOI] [PubMed] [Google Scholar]
  • [6].Jia R, Li N, Xu B, Zhang W, Gu X, Ma Z. Incidence, influencing factors, and prognostic impact of intraoperative massive blood loss in adolescents with neuromuscular scoliosis: A Strobe-compliant retrospective observational analysis. Medicine (Baltimore) 2017;96. 10.1097/MD.0000000000006292. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Baker CE, Marvi T, Austin TM, Payne S, Mignemi ME, Gailani D, et al. Dilutional coagulopathy in pediatric scoliosis surgery: A single center report. Paediatr Anaesth 2018;28:974–81. 10.1111/pan.13488. [DOI] [PubMed] [Google Scholar]
  • [8].Drucker NA, Wang SK, Newton C. Pediatric trauma-related coagulopathy: Balanced resuscitation, goal-directed therapy and viscoelastic assays. Semin Pediatr Surg 2019;28:61–6. 10.1053/j.sempedsurg.2019.01.011. [DOI] [PubMed] [Google Scholar]
  • [9].Leeper CM, Neal MD, McKenna CJ, Gaines BA. Trending Fibrinolytic Dysregulation: Fibrinolysis Shutdown in the Days After Injury Is Associated With Poor Outcome in Severely Injured Children. Ann Surg 2017;266:508–15. 10.1097/SLA.0000000000002355. [DOI] [PubMed] [Google Scholar]
  • [10].Yuan L, Zeng Y, Chen ZQ, Zhang XL, Mai S, Song P, et al. Efficacy and safety of antifibrinolytic agents in spinal surgery: A network meta-analysis. Chin Med J (Engl) 2019;132:577–88. 10.1097/CM9.0000000000000108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Kozek-Langenecker SA, Afshari A, Albaladejo P, Santullano CAA, De Robertis E, Filipescu DC, et al. Management of severe perioperative bleeding: guidelines from the European Society of Anaesthesiology. Eur J Anaesthesiol 2013;30:270–382. 10.1097/EJA.0b013e32835f4d5b. [DOI] [PubMed] [Google Scholar]
  • [12].Pabinger I, Fries D, Schöchl H, Streif W, Toller W. Tranexamic acid for treatment and prophylaxis of bleeding and hyperfibrinolysis. Wien Klin Wochenschr 2017;129:303–16. 10.1007/s00508-017-1194-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Goobie SM, Zurakowski D, Glotzbecker MP, McCann ME, Hedequist D, Brustowicz RM, et al. Tranexamic Acid Is Efficacious at Decreasing the Rate of Blood Loss in Adolescent Scoliosis Surgery: A Randomized Placebo-Controlled Trial. J Bone Jt Surg 2018;100:2024–32. 10.2106/JBJS.18.00314. [DOI] [PubMed] [Google Scholar]
  • [14].Faraoni D, Goobie SM. The efficacy of antifibrinolytic drugs in children undergoing noncardiac surgery: a systematic review of the literature. Anesth Analg 2014;118:628–36. 10.1213/ANE.0000000000000080. [DOI] [PubMed] [Google Scholar]
  • [15].Goobie SM, Meier PM, Pereira LM, McGowan FX, Prescilla RP, Scharp LA, et al. Efficacy of Tranexamic Acid in Pediatric Craniosynostosis SurgeryA Double-blind, Placebo-controlled Trial. Anesthesiol J Am Soc Anesthesiol 2011;114:862–71. 10.1097/ALN.0b013e318210fd8f. [DOI] [PubMed] [Google Scholar]
  • [16].Gausden EB, Brusalis CM, Qudsi RA, Swarup I, Fu M, Dodwell E, et al. Efficacy of antifibrinolytics in pediatric orthopedic surgery: a systematic review and meta-analysis. J Pediatr Orthop Part B 2020;29:97–104. 10.1097/BPB.0000000000000599. [DOI] [PubMed] [Google Scholar]
  • [17].Goobie SM, Faraoni D. Tranexamic acid and perioperative bleeding in children: what do we still need to know? Curr Opin Anaesthesiol 2019;32:343–52. 10.1097/ACO.0000000000000728. [DOI] [PubMed] [Google Scholar]
  • [18].Johnson DJ, Johnson CC, Goobie SM, Nami N, Wetzler JA, Sponseller PD, et al. High-dose Versus Low-dose Tranexamic Acid to Reduce Transfusion Requirements in Pediatric Scoliosis Surgery. J Pediatr Orthop 2017;37:e552–7. 10.1097/BPO.0000000000000820. [DOI] [PubMed] [Google Scholar]
  • [19].Shrestha IK, Ruan T-Y, Lin L, Tan M, Na X-Q, Qu Q-C, et al. The efficacy and safety of high-dose tranexamic acid in adolescent idiopathic scoliosis: a meta-analysis. J Orthop Surg 2021;16:53. 10.1186/s13018-020-02158-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Stricker PA, Gastonguay MR, Singh D, Fiadjoe JE, Sussman EM, Pruitt EY, et al. Population pharmacokinetics of ϵ-aminocaproic acid in adolescents undergoing posterior spinal fusion surgery. Br J Anaesth 2015;114:689–99. 10.1093/bja/aeu459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Kannan S, Meert KL, Mooney JF, Hillman-Wiseman C, Warrier I. Bleeding and coagulation changes during spinal fusion surgery: a comparison of neuromuscular and idiopathic scoliosis patients. Pediatr Crit Care Med J Soc Crit Care Med World Fed Pediatr Intensive Crit Care Soc 2002;3:364–9. 10.1097/00130478-200210000-00007. [DOI] [PubMed] [Google Scholar]
  • [22].Jaramillo S, Montane-Muntane M, Gambus PL, Capitan D, Navarro-Ripoll R, Blasi A. Perioperative blood loss: estimation of blood volume loss or haemoglobin mass loss? Blood Transfus Trasfus Sangue 2020;18:20–9. 10.2450/2019.0204-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Jaramillo S, Montane‐Muntane M, Capitan D, Aguilar F, Vilaseca A, Blasi A, et al. Agreement of surgical blood loss estimation methods. Transfusion (Paris) 2019;59:508–15. 10.1111/trf.15052. [DOI] [PubMed] [Google Scholar]
  • [24].Minhas SV., Chow I, Bosco J, Otsuka NY. Assessing the Rates, Predictors, and Complications of Blood Transfusion Volume in Posterior Arthrodesis for Adolescent Idiopathic Scoliosis. Spine 2015;40:1422–30. 10.1097/BRS.0000000000001019. [DOI] [PubMed] [Google Scholar]
  • [25].Sharma V, Katznelson R, Jerath A, Garrido-Olivares L, Carroll J, Rao V, et al. The association between tranexamic acid and convulsive seizures after cardiac surgery: a multivariate analysis in 11 529 patients. Anaesthesia 2014;69:124–30. 10.1111/anae.12516. [DOI] [PubMed] [Google Scholar]
  • [26].Martin K, Breuer T, Gertler R, Hapfelmeier A, Schreiber C, Lange R, et al. Tranexamic acid versus ɛ-aminocaproic acid: efficacy and safety in paediatric cardiac surgery. Eur J Cardiothorac Surg 2011;39:892–7. 10.1016/j.ejcts.2010.09.041. [DOI] [PubMed] [Google Scholar]
  • [27].Shapiro F, Sethna N. Blood loss in pediatric spine surgery. Eur Spine J 2004;13:S6–17. 10.1007/s00586-004-0760-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [28].Meert KL, Kannan S, Mooney JF. Predictors of red cell transfusion in children and adolescents undergoing spinal fusion surgery. Spine 2002;27:2137–42. 10.1097/00007632-200210010-00012. [DOI] [PubMed] [Google Scholar]
  • [29].Yoshihara H, Yoneoka D. Predictors of allogeneic blood transfusion in spinal fusion for pediatric patients with idiopathic scoliosis in the United States, 2004–2009. Spine 2014;39:1860–7. 10.1097/BRS.0000000000000530. [DOI] [PubMed] [Google Scholar]
  • [30].Yoshihara H, Yoneoka D. National Trends in Spinal Fusion for Pediatric Patients With Idiopathic Scoliosis: Demographics, Blood Transfusions, and In-hospital Outcomes. Spine 2014;39:1144–50. 10.1097/BRS.0000000000000354. [DOI] [PubMed] [Google Scholar]
  • [31].Halanski MA, Cassidy JA, Hetzel S, Reischmann D, Hassan N. The Efficacy of Amicar Versus Tranexamic Acid in Pediatric Spinal Deformity Surgery: A Prospective, Randomized, Double-Blinded Pilot Study. Spine Deform 2014;2:191–7. 10.1016/j.jspd.2014.02.001. [DOI] [PubMed] [Google Scholar]
  • [32].Neilipovitz DT, Murto K, Hall L, Barrowman NJ, Splinter WM. A randomized trial of tranexamic acid to reduce blood transfusion for scoliosis surgery. Anesth Analg 2001;93:82–7. [DOI] [PubMed] [Google Scholar]
  • [33].Lavoie J. Blood transfusion risks and alternative strategies in pediatric patients. Paediatr Anaesth 2011;21:14–24. 10.1111/j.1460-9592.2010.03470.x. [DOI] [PubMed] [Google Scholar]
  • [34].Verma K, Errico T, Diefenbach C, Hoelscher C, Peters A, Dryer J, et al. The relative efficacy of antifibrinolytics in adolescent idiopathic scoliosis: a prospective randomized trial. J Bone Jt Surg 2014;96:e80. 10.2106/JBJS.L.00008. [DOI] [PubMed] [Google Scholar]
  • [35].Halanski MA, Cassidy JA, Hetzel S, Reischmann D, Hassan N. The efficacy of amicar versus tranexamic acid in pediatric spinal deformity surgery: A prospective, randomized, double-blinded pilot study. Spine Deform 2014;2:191–7. 10.1016/j.jspd.2014.02.001. [DOI] [PubMed] [Google Scholar]
  • [36].Stutz CM, OʼRear LD, O’Neill KR, Tamborski ME, Crosby CG, Devin CJ, et al. Coagulopathies in Orthopaedics: Links to Inflammation and the Potential of Individualizing Treatment Strategies. J Orthop Trauma 2013;27:236–41. 10.1097/BOT.0b013e318269b782. [DOI] [PubMed] [Google Scholar]
  • [37].Lecker I, Wang D, Whissell PD, Avramescu S, Mazer CD, Orser BA. Tranexamic acid–associated seizures: Causes and treatment. Ann Neurol 2016;79:18–26. 10.1002/ana.24558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [38].Miszta A, Huskens D, Donkervoort D, Roberts MJM, Wolberg AS, de Laat B. Assessing Plasmin Generation in Health and Disease. Int J Mol Sci 2021;22:2758. 10.3390/ijms22052758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [39].Goobie SM, Cladis FP, Glover CD, Huang H, Reddy SK, Fernandez AM, et al. Safety of antifibrinolytics in cranial vault reconstructive surgery: a report from the pediatric craniofacial collaborative group. Pediatr Anesth 2017;27:271–81. 10.1111/pan.13076. [DOI] [PubMed] [Google Scholar]
  • [40].Kramer M, Drexler M, Herman A, Kalimian T, Klassov Y, Nasser LA. Use of Intraoperative Tranexamic Acid and Wound Complications in Spine Surgery: A Retrospective Cohort Study. Asian Spine J 2020;14:639–46. 10.31616/asj.2019.0235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [41].Hassan N, Halanski M, Wincek J, Reischman D, Sanfilippo D, Rajasekaran S, et al. Blood management in pediatric spinal deformity surgery: review of a 2-year experience. Transfusion (Paris) 2011;51:2133–41. 10.1111/j.1537-2995.2011.03175.x. [DOI] [PubMed] [Google Scholar]
  • [42].Brenn BR, Theroux MC, Dabney KW, Miller F. Clotting parameters and thromboelastography in children with neuromuscular and idiopathic scoliosis undergoing posterior spinal fusion. Spine 2004;29:E310—4. 10.1097/01.brs.0000132513.88038.64. [DOI] [PubMed] [Google Scholar]
  • [43].Stutz CM, O’Rear LD, O’Neill KR, Tamborski ME, Crosby CG, Devin CJ, et al. Coagulopathies in orthopaedics: Links to inflammation and the potential of individualizing treatment strategies. J Orthop Trauma 2013;27:236–41. 10.1097/BOT.0b013e318269b782. [DOI] [PubMed] [Google Scholar]
  • [44].Caubet J-F, Emans JB, Smith JT, vanBosse H, Ramirez N, Flynn J, et al. Increased Hemoglobin Levels in Patients With Early Onset Scoliosis: Prevalence and Effect of a Treatment With Vertical Expandable Prosthetic Titanium Rib (VEPTR). Spine 2009;34:2534–6. 10.1097/BRS.0b013e3181bd0fc4. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Data File (.doc, .tif, pdf, etc.)
NIHMS1741824-supplement-Supplemental_Data_File___doc___tif__pdf__etc__.docx (12.6KB, docx)

RESOURCES