The authors have declared that no competing interests exist.
Conceived and designed the experiments: HL. Performed the experiments: BY. Designed the software: PY. Analyzed the data: DW. Contributed reagents/materials/analysis tools: XH. Wrote the paper: CZ.
Tranexamic acid (TXA) is well-established as a versatile oral, intramuscular, and intravenous (IV) antifibrinolytic agent. However, the efficacy of IV TXA in reducing perioperative blood transfusion in spinal surgery is poorly documented.
We conducted a meta-analysis of randomized controlled trials (RCTs) and quasi-randomized (qi-RCTs) trials that included patients for various spinal surgeries, such as adolescent scoliosis surgery administered with perioperative IV TXA according to Cochrane Collaboration guidelines using electronic PubMed, Cochrane Central Register of Controlled Trials, and Embase databases. Additional journal articles and conference proceedings were manually located by two independent researchers.
Totally, nine studies were included, with a total sample size of 581 patients. Mean blood loss was decreased in patients treated with perioperative IV TXA by 128.28 ml intraoperatively (ranging from 33.84 to 222.73 ml), 98.49 ml postoperatively (ranging from 83.22 to 113.77 ml), and 389.21 ml combined (ranging from 177.83 to 600.60 ml). The mean volume of transfused packed cells were reduced by 134.55 ml (ranging 51.64 to 217.46) (95% CI; P = 0.0001). Overall, the number of patients treated with TXA who required blood transfusions was lower by 35% than that of patients treated with the comparator and who required blood transfusions (RR 0.65; 95% CI; 0.53 to 0.85; P<0.0001, I2 = 0%). A dose-independent beneficial effect of TXA was observed, and confirmed in subgroup and sensitivity analyses. A total of seven studies reported DVT data. The study containing only a single DVT case was not combined.
The blood loss was reduced in spinal surgery patients with perioperative IV TXA treatment. Also the percentage of spinal surgery patients who required blood transfusion was significantly decreased. Further evaluation is required to confirm our findings before TXA can be safely used in patients undergoing spine surgery.
Server perioperative blood loss during procedures is a major concern during procedures, such as spinal surgery; therefore, allogeneic blood transfusion is a common requirement
A variety of contemporary blood-conservation techniques have been used to reduce exposure to allogeneic blood, including controlled hypotension, regional anesthesia, autologous blood transfusion, intraoperative blood salvage, and administration of various intravenous (IV), intramuscular, and oral medications
Based on a recent comprehensive evaluation and meta-analysis of the effects of antifibrinolytic drug on both bleeding and transfusion, it could significantly reduce blood loss, thus reduce the need for allogeneic red blood cell transfusions in surgical patients
Fibrinolysis represents a unique challenge in spinal surgery and postsurgical care. It has been suggested that antifibrinolytic drugs, such as perioperative IV TXA, should be applied more routinely in clinical spinal surgeries. Therefore, we conducted a systematic review and meta-analysis of the efficacy of TXA in reducing blood loss and exposure to perioperative blood transfusion during spinal. We also further examined the association of IV TXA with complication rates.
A meta-analysis and systematic review was conducted according to a predefined guidelines provided by the Cochrane Collaboration (2008)
We performed a comprehensive search based on multi-database electronic and manual literature searches to identify relevant controlled trials (RCTs) and quasi-randomized trials (qi-RCTs) published between December 1966 and September 2012 involving patients undergoing a variety of spinal surgery, including adolescent spinal scoliosis surgery and other surgery types. Electronic searches were conducted in database including the PubMed, Cochrane Central Register of Controlled Trials, and Embase databases without language restrictions. The search terms included “antifibrinolytics”, “tranexamic acid”, “cyklokapron”, “aprotinin”, “trasylol”, “epsilon aminocaproic acid”, “amicar” and “randomized controlled trials”. Manual searches were conducted only in English using the Orthopedics China Biological Medicine database.
The studies were included if they met the following criteria: (i) participants underwent spinal surgery; (ii) interventions included administration of perioperative IV TXA; (iii) results of a placebo (control) group were reported; (iv) reported outcomes including intraoperative, postoperative, and total blood loss (primary outcomes); and (v) reported outcomes included the number of patients receiving allogeneic blood transfusion, transfusion packed cells volumes, and deep venous thrombosis (DVT) incidence (secondary outcomes).
Studies reporting the results of oral or intramuscular TXA interventions were excluded. Non-randomized studies and studies reporting the results of spinal revision were also excluded. In this study, the allocation sequence of RCTs was considered to be unpredictable (e.g. selection by random computer generated sequences), while that of qi-RCTs was considered to be predictable (e.g. selection by medical record numbers or birth dates).
The initial electronic databases searches to identify potential studies for inclusion based on title and abstract information were performed by two independent authors (BaoHui Yang and HaoPeng Li). Complete study reports were assessed for inclusion independently by both authors. In cases of insufficient data, authors were contacted for information and clarification. Final determination of inclusion was made by the senior author (Xijing He). If any author presented an objection to inclusion because studies do not meet the inclusion criteria, the study was excluded from the analysis. References and data for each included study were carefully cross-checked to ensure no overlapping data was presented, thereby to ensure the integrity of the meta-analysis results.
The general characteristics, treatment or intervention types, and outcomes were recorded for each study. General characteristics included study design, publication date, participant demographics and number, geographic location, and interventions. For all included studies, treatment and interventions were recorded, including IV TXA dose regimen and timing, anesthesia method(s), and surgical intervention(s). Measured outcomes included the primary and secondary outcomes as described previously. In addition, all adverse outcomes were recorded. For studies without adverse outcomes, author(s) were contacted via e-mail for confirmation or more information regarding adverse events, if necessary.
The included studies were assessed for risk of bias by 2 independent researchers and a managing reviewer according to the Cochrane guidelines
Meta-analysis on the data, including effects and outcomes of treatments, was performed using the Mantel-Haenszel method with Review Manager software (RevMan version 5.0, Cochrane Collaboration, Germany). For continuous data, such as blood loss, mean ± standard deviation (mean ± SD) was used to calculate the weighted mean difference (WMD) and 95% confidence interval (CI). For Dichotomous data, relative risk (RR) and 95% CI were applied.
Statistical heterogeneity was assessed using the value of I2 and the result of the chi-squared test A p-value <0.1 and an I2 value >50% were considered suggestive of statistical heterogeneity, prompting a random effects modeling estimate. Otherwise, a fixed effects approach was used. However, a non-significant chi-squared test result (a p-value ≥0.1 and an I2 value ≤50%) only indicated a lack of evidence for heterogeneity, but not implied necessarily homogeneity, as there may have been insufficient power to be able to detect heterogeneity.
If any heterogeneity was observed, the causes of heterogeneity was first analyzed and then subjected to sub-group treatment. If the statistical heterogeneity cannot be eliminated, a random effect model was used for the combined analysis of the studies to examine the clinical consistency.
The initial search identified a total of 990 articles from the electronic database (Pubmed: n = 276; Cochrane: n = 443; Embase: n = 269) and manual search (Orthopedics China Bio Med: n = 2). After exclusion of 585 studies containing overlapping data or appearing in more than one database, 445 studies remained. After screening the titles and abstracts, 397 studies that do not meet the inclusion criteria were further excluded, and 31 additional studies without specific information in spinal surgery also were excluded. In the remained 17 studies, 8 studies were excluded based on the exclusion criteria provided. Finally, 9 studies (5 RCTs and 2 qi-RCTs) that met all inclusion criteria were included in our study, including 7 studies published in English, 1 study in Chinese, and 1 study in Korean. These 9 studies included a total of 581 patients. A detailed flow chat for selection is shown in
For all included studies, the methodological quality represented a minimal risk of bias-based error or uncertainty. All studies provided detailed information on the randomization techniques applied, including randomization generated by computer programs
review authors' judgments about each risk of bias item for each included study. + is “yes”, − is “no”, ? is “unclear”.
As shown in
For each study, TXA was administered by IV, though variant dosages (ranging form 10–100 mg/kg) and delivery timings were applied. In 2 studies, a single IV bolus was administered preoperatively
DVT screening was conducted by either clinical assessment
A total of 8 studies (n = 501; 253 experimental and 248 control patients)
The black diamond signifies that the mean difference is in favour of TXA. The size of each square depends on the weight of each study. A green square is given to continuous outcomes.
A total of 5 studies (n = 341; 170 experimental and 171 control patients)
The black diamond signifies that the mean difference is in favour of TXA. The size of each square depends on the weight of each study. A green square is given to continuous outcomes.
A total of 6 studies (n = 397; 199 experimental and 198 control patients)
The black diamond signifies that the mean difference is in favour of TXA. The size of each square depends on the weight of each study. A green square is given to continuous outcomes.
A total of 7 studies (n = 461; 233 experimental and 228 control patients)
The black diamond signifies that the mean difference is in favour of TXA. The size of each square depends on the weight of each study. A blue square is given to dichotomous outcomes.
A total of 6 studies (n = 342; 171 experimental and 171 control patients)
A total of 7 studies (n = 479) reported DVT data. One study with only a single DVT case was not included
Heterogeneity was first analyzed and then subjected to subgroup treatment. Based on heterogeneity data, subgroup analysis was designed for anesthesia type, IV TXA dose regime and timing, spinal surgery delivery, transfusion triggers, and surgical methods for studies containing sufficient and complete data sets for each of these variables. However, considering of the incomplete data, the included studies were finally divided by TXA loading drug dose as low (≤15 mg/kg) and high (>15 mg/kg) doses. Subgroup analysis revealed that both high
Sensitivity analysis showed that transfusion rates in patients treated with perioperative IV TXA were lower than that in control patients (RR 0.53; 95% CI 0.34 to 0.85; P = 0.008) (
Blood transfusion rates were also used to generate the funnel plot analysis of publication bias (
Funnel plot to assess publication for the most frequently reported outcome –blood transfusion rate.
Blood loss both during and after spinal surgery, a major concern, makes it necessary to perform transfusion, which may present more risks to patients. Thus, a better understanding of the clinical effects of hemostatic techniques is required. The synthetic antifibrinolytic agent, TXA, has been demonstrated to reduce the need for blood transfusions, thereby prevent adverse complications, such as transfusion reactions and spinal epidural hematoma formation. Evaluation on the dose- and time-related efficacy, as well as a widespread implementation of current hemostatic techniques may reduce the need for transfusion procedures in clinical spinal surgery patients and improve overall patient outcomes.
Transfusion is common in patients undergoing spinal surgery. Extensive evidence has demonstrated that patients undergoing such surgery and receiving transfusion are at risk of numerous moderate and severe complications. Thus, a safe and effective intervention to reduce the number of blood units to these patients during and immediately following surgery is required. TXA has been successfully applied to treat major hemorrhaging in adult spinal surgery
TXA has been demonstrated to be effective in numerous spinal surgeries in adults, adolescents, and children, suggesting that it might be a potential treatment for various surgical procedures. In scoliosis correction surgery, TXA was shown to be effective when used along with other blood conservation measures
A systematic review and meta-analysis
In order to fully investigate the clinical effects associated with TXA administration, the included studies were selected independently of whether they supported the TXA use, including 2 studies that did not support the routine use of TXA in spinal surgery. In a double-blind RCT of 76 patients in spinal fixation surgery, TXA was administered upon induction of anesthesia for 10 min, followed by IV infusion of TXA at a rate of 1 mg/kg/h in 38 patients, and an equivalent protocol of normal saline in 38 patients as control
A recent meta-analysis has assessed the effect of TAX on blood transfusion, thromboembolic events, and mortality in surgical patients
Current study has been focused on the effects of perioperative administration of TXA independent of dosage and timing. Further exploration of intraoperative TXA administration is required to fully assess the potential benefits of widespread clinical administration. In one retrospective study, a 50% reduction in transfusion requirements were observed in patients treated with higher intraoperative doses of TXA
In summary, perioperative IV TXA significantly reduces intra-operative blood loss and transfusion requirements for spinal surgery patients. Furthermore, TXA administration was not linked with any significant increase in complication rates, including DVT. Better post-operative outcomes are also achieved by the reduced total blood loss achieved by TXA administration. Considering the significant heterogeneity and relatively small number of studies included, further larger studies are needed to confirm the findings, as well as to determine the optimal TXA dosages and administration timing for spinal surgery patients.
Table S1. Characteristics of Included Randomized Controlled Trials and quasi-RCTs. Table S2. Subgroup analysis of outcome for low doses of perioperative IV TXA. Table S3. Subgroup analysis of outcomes for high doses of perioperative IV TXA. Table S4. Sensitivity analyses for excluded qi-RCTs.
(DOC)