Keywords
chronic subdural hematoma - craniotomy - subdural hematoma - tranexamic acid
Introduction
Chronic subdural hematoma (CSDH) commonly occurs in elderly patients, and in the aging
population, the incidence is expected to increase.[1] The burr hole craniotomy (BC) is the standard management, which shows promising
outcomes. However, the risk of hematoma recurrence varies by 10 to 29% in 1 to 8 weeks.[2]
[3] However, since the development of the endovascular technique, middle meningeal artery
embolization demonstrated high efficacy as an alternative or adjunctive approach in
the management of CSDH.[4]
[5] Regarding the promising results of different surgery techniques and patient selection,[6] the major concern about the priority of modalities and patient selection is still
not proven.
In addition, nonsurgical techniques and drug prescriptions, including dexamethasone,[7]
[8] angiotensin-converting enzyme inhibitors,[9] tranexamic acid (TXA),[10]
[11] and herbal medicine,[12] have been reported in the literature. It is still unknown which of these drugs are
safe and suitable for the elderly, given their high incidence of frailty to be prescribed
with multiple medications. In the CSDH cavity, inflammatory mediators stimulate vascular
permeability and release tissue plasminogen activators from endothelial cells, increasing
plasminogen conversion to plasmin.[13] TXA reversibly binds to plasminogen, preventing the binding of plasmin to fibrin
and degradation of fibrin, subsequently.[14] Although previous studies have reported a preventative effect of oral TXA for CSDH,[15] hyperfibrinolysis contributes to continuous bleeding in the cavity and the recurrence
of hematoma.[16] This study aimed to investigate the safety and efficacy of TXA as an adjunct option
of surgery on CSDH.
Materials and Methods
Protocol and Registration
This systematic review and meta-analysis are formulated by the Preferred Reporting
Items for Systematic Reviews and Meta-Analyses guidelines and the Cochrane Handbook
for Systematic Reviews of Interventions. The study has not been registered in either
PROSPERO or another systematic reviews database.
Database and Search Strategy
A comprehensive literature search, spanning from inception to January 2024, was performed
across the following databases: PubMed, SCOPUS, EMBASE, and Web of Science from inception
to September 21, 2023. The following search terms, “Tranexamic acid,” “recurrence,”
and “Chronic Subdural Hematoma,” were combined using Boolean operators and truncation
([Table 1]). The search was limited to English-language or available English translation publications.
Table 1
Search strategies for various databases
|
Database
|
Search strategy
|
Number
|
|
PubMed
|
#1 “tranexamic acid”[MeSH Terms] OR (“tranexamic”[All Fields] AND “acid”[All Fields])
OR “tranexamic acid”[All Fields]
#3 “Hematoma, Subdural”[Mesh] OR “Hematoma, Subdural, Intracranial”[Mesh] OR “Hematoma,
Subdural”[Mesh] OR “subdural hematoma”[tiab] OR “subdural haematoma”[tiab] OR “subdural
hematoma”[tiab] OR “haematoma, subdural”[tiab] OR “haematoma, subdural”[tiab] OR “haemorrhage,
subdural”[tiab] OR “hematoma, subdural”[tiab] OR “hemorrhage, subdural”[tiab] OR “subdural
bleeding”[tiab] OR “subdural haematoma”[tiab] OR “subdural haemorrhage”[tiab] OR “subdural
hemorrhage”[tiab] OR “subdural hematoma”[tiab]
#4 #1 AND #2 AND #3
|
54
|
|
SCOPUS
|
#1 TITLE-ABS-KEY (“tranexamic acid” OR (“tranexamic” AND “acid”) OR “tranexamic acid”)
#3 TITLE-ABS-KEY (“subdural hematoma” OR “subdural haematoma” OR “subdural hematoma”
OR “haematoma, subdural” OR “haemorrhage, subdural” OR “hematoma, subdural” OR “hemorrhage,
subdural” OR “subdural bleeding” OR “subdural haematoma” OR “subdural haemorrhage”
OR “subdural hemorrhage” OR “subdural hematoma”)
#4 #1 AND #2 AND #3
|
158
|
|
Embase
|
#1 'tranexamic acid'/exp OR 'tranexamic acid':ti,ab OR ('tranexamic':ti,ab AND 'acid':ti,ab)
#2 'subdural hematoma'/exp OR 'subdural haematoma':ti,ab OR 'subdural hematoma':ti,ab
OR 'haematoma, subdural':ti,ab OR 'haematoma, subdural, intracranial':ti,ab OR 'hemorrhage,
subdural':ti,ab OR 'hematoma, subdural':ti,ab OR 'hematoma, subdural, acute':ti,ab
OR 'hematoma, subdural, intracranial':ti,ab OR 'hemorrhage, subdural':ti,ab OR 'subdural
bleeding':ti,ab OR 'subdural haematoma':ti,ab OR 'subdural hemorrhage':ti,ab OR 'subdural
hemorrhage':ti,ab OR 'subdural hematoma':ti,ab
#4 #1 AND #2 AND #3
|
184
|
|
WOS
|
#1 TI= (“tranexamic acid” OR (“tranexamic” AND “acid”) OR “tranexamic acid”)
#2 TI= (“subdural hematoma” OR “subdural haematoma” OR “subdural hematoma” OR “haematoma,
subdural” OR “haematoma, subdural” OR “haemorrhage, subdural” OR “hematoma, subdural”
OR “hemorrhage, subdural” OR “subdural bleeding” OR “subdural haematoma” OR “subdural
haemorrhage” OR “subdural hemorrhage” OR “subdural hematoma”)
#4 #1 AND #2 AND #3
|
17
|
Abbreviations: MeSH: Medical Subject Headings; WOS, Web of Science.
Note: 219 duplicates removed. September 21, 2023.
The inclusion and exclusion criteria were applied to screen the titles and abstracts
of the identified studies. Full-text assessments followed for potentially eligible
studies, along with manual screening of reference lists of these articles to identify
other relevant studies. The cohort studies, randomized controlled trials, or propensity-matched
cohorts that administered TXA were included. Conference data, review papers, case
reports with less than 10 patients, studies where TXA was used as the primary medical
management, and study protocols were excluded. Of note, papers including patients
with recurrent CSDH, such as dural arteriovenous fistula, were excluded from the final
investigation.
Study Selection
Two independent reviewers performed complete screening to ensure the accuracy and
study relevance. Initially, the title and abstract of the selected research were examined
to identify and remove any unrelated sources. Then, the full texts of the papers were
comprehensively reviewed, facilitating subsequent assessment by two authors meticulously.
The manuscript that met the predetermined inclusion criteria of the study was chosen
for data extraction. Additionally, to reduce the effect of the viewpoint of the two
authors, a third reviewer resolved the disparities. A thorough examination of the
references cited within the selected papers was conducted to identify more relevant
papers.
Data Extraction
Two reviewers conducted the data extraction process. The information included the
author's name, study period, study designation, number of patients, demographic data,
hematoma characteristics, complications, mortality, hematoma volume, and follow-up
period. The primary outcomes were hematoma recurrence, reoperation rate, and hematoma
reduction, and the secondary outcomes were thrombosis, serious adverse events (SAEs),
and mortality.
Statistical Analysis
The statistical analysis was performed with STATA V.17. The primary outcome, the rate
of recurrence, and the secondary outcome, the rate of thrombosis, were pooled using
95% confidence intervals (CIs) and a random effects model. Subgroup analysis was conducted
to explore the outcome based on the study type (randomized control trial and propensity-matched
cohorts and case series with more than 10 patients). Heterogeneity was assessed using
the I
2 test, and an I
2 < 30% was defined as low heterogeneity. All statistical tests were two-sided; a p-value of < 0.05 was considered statistically significant. Means are reported as mean
(95% CI). Egger's test was used to assess publication bias.
Results
Study Selection Process
The initial search strategy yielded 413 articles in the PubMed, Embase, Scopus, and
Web of Science databases. A total of 219 articles were deduplicated, and the remaining
194 articles underwent title/abstract screening, of which 172 articles were removed.
A total of 22 articles underwent full-text screening, and studies that did not fully
meet the inclusion criteria were excluded (n = 12). Finally, we included 10 studies in the systematic review. Five of the included
studies were randomized control trials.[17]
[18]
[19]
[20]
[21] The remaining two studies were propensity-matched cohorts[22]
[23] and three retrospective study.[10]
[11]
[24] All the included studies were recently published, encompassing the time frame from
2012 to 2023. [Fig. 1] outlines the Preferred Reporting Items for Systematic Reviews and Meta-Analyses
flow diagram of the study selection process.
Fig. 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart
of the study selection process.
Baseline Characteristics
This study included 14,836 patients with subdural hematoma who underwent bur hole
hematoma evacuation and received TXA. A total of 7,450 patients only used surgery
to treat CSDH (control group), and 7,386 patients received TXA as an adjuvant treatment
for the surgical intervention (TXA group). The mean age of the patients was reported
in nine studies. The median follow-up of patients was 6 months (range 1–12 months).
The duration of TXA administration for patients reported in five studies varied from
21 days to 6 moths.[11]
[17]
[19]
[20]
[21] Most studies utilized 750 mg/day of TXA daily,[10]
[17]
[21]
[22]
[23]
[24] and the remaining studies used 650 mg/day,[11] 1,000 mg/day of TXA,[18]
[19] and 1,500 mg/day.[20]
[Table 2] represents the baseline characteristics of the included study.
Table 2
Included studies summary
|
Author, year
|
Study type
|
No of patients (control)
|
% Male (control)
|
Age (control)
|
Follow-up (mo)
|
|
de Paula et al, 2023[17]
|
RCT
|
50 (26)
|
62 (61.5)
|
75.8 ± 11.8 (72.6 ± 11.9)
|
9.4
|
|
Kageyama et al, 2013[10]
|
Retrospective
|
21
|
12
|
72.0 (71.5)
|
10
|
|
Miyakoshi et al, 2023[22]
|
PSM
|
930 (465)
|
65.1 (65.1)
|
81.3 ± 7.3 (81. 1 ± 6.9)
|
6
|
|
Perel et al, 2012[18]
|
RCT
|
270 (137)
|
228 (85.4)
|
36 ± 14 (37 ± 14)
|
1
|
|
Wan et al, 2020[19]
|
RCT
|
90 (49)
|
66.7 (48.9)
|
69.57 ± 13.69 (72.02 ± 11.79)
|
6
|
|
Shibahashi et al,2022[23]
|
PSM
|
13,128 (6,564)
|
4,523 (68.9)
|
Not specified
|
12
|
|
Tanweer et al, 2016[11]
|
Retrospective
|
14
|
12
|
56.4 ± 16
|
3
|
|
Workewych, 2018[20]
|
RCT
|
24 (13)
|
41.6 (46.1)
|
70.18 ± 12.03 (70.85 ± 9.31)
|
2–3
|
|
Yamada and Natori, 2020[21]
|
RCT
|
154 (82)
|
64.9 (69.5)
|
78.2 ± 9.8 (78.8 ± 10.8)
|
3
|
|
Yang et al, 2023[24]
|
Retrospective
|
155 (114)
|
77.4 (76.3)
|
72.0 (71.5)
|
10
|
Abbreviations: PSM, propensity score matching; RCT, randomized clinical trial.
Primary Outcome
Recurrence
In the five studies that reported hematoma recurrence, the pooled recurrence rate
in the TXA group was found to be 0.02 (95% CI [0.00–0.04], I
2: 6.16%, p-value = 0.32) and the control group 0.08 (95% CI [0.05–0.11], I
2: 0.01%, p-value: 0.66). Test of group difference revealed that adjuvant TXA to surgical management
was associated with significant decrease in recurrence rate (p-value < 0.001) ([Fig. 2]).
Fig. 2 The recurrence rate between adjuvant tranexamic acid (TXA) to surgery (TXA group)
and surgical management only (control group).
Reoperation
Three studies included the required reoperation rate (two cohort studies and one randomized
clinical trial [RCT]). The overall reoperation rate in the TXA group was 0.05 (95%
CI [0.00–0.10], I
2: 92.59%, p-value < 0.001) and in the control group was 0.12 (95% CI [0.03–0.20], I
2: 91.7%, p-value < 0.001). Test of group difference showed no significant difference between
adjuvant TXA compared with the control group (p-value: 0.18) ([Fig. 3]).
Fig. 3 The reoperation rate between adjuvant tranexamic acid (TXA) to surgery (TXA group)
and surgical management only (control group).
Hematoma Volume Reduction
Two RCT studies evaluated the hematoma reduction, the overall SDH volume was –0.36
(95% CI [–0.66 to –007], I
2: 98%, p-value = 0.98), lower in the TXA group than the control group, which was significantly
different (p-value < 0.02) ([Fig. 4]).
Fig. 4 The hematoma reduction between adjuvant tranexamic acid (TXA) to surgery (TXA group)
and surgical management only (control group).
Secondary Outcome
Mortality Rate
Three studies evaluated the mortality rate (two RCTs and one cohort study), the mortality
rate in the TXA group was 0.06 (95% CI [0.03–0.10], I
2: 49.70%, p-value = 0.15), and in the control group was 0.07 (95%CI [–0.03 to 0.18], I
2: 98.87%, p-value < 0.001). The overall mortality rate was reduced in the TXA group compared
with the control group but the difference was not statistically significant (p-value: 0.87) ([Fig. 5]).
Fig. 5 The comparison of mortality rate between adjuvant tranexamic acid (TXA) to surgery
(TXA group) and surgical management only (control group).
Serious Adverse Events
Three RCTs categorized the SAEs, including new onset or growth of the intracranial
hemorrhage, the need for neurosurgery, or death. The SAE in the TXA group was found
to be 0.20 (95% CI [–0.05 to 0.40], I
2: 96.25%, p-value < 0.001) and 0.19 (95% CI [–0.19 to 0.57], I
2: 100%, p-value < 0.001) in the control group. Test of group difference was shown and no significant
difference was present between the TXA and control group ([Fig. 6]).
Fig. 6 The forest plot of serious adverse events between adjuvant tranexamic acid (TXA)
to surgery (TXA group) and surgical management only (control group). Complications
were not significantly different between the two groups (p-Value: 0.96).
Fig. 7 The forest plot of thromboembolic complication rate between adjuvant tranexamic acid
(TXA) to surgery (TXA group) and surgical management only (control group).
Thromboembolic Complications
Four studies (two RCTs and two cohort studies) evaluated thromboembolic complications
(TCs). TC in the TXA group was found to be 0.05 (95% CI [–0.02 to 0.11], I
2: 100%, p-value < 0.001) and 0.04 (95% CI [–0.04 to 0.12], I
2: 100%, p-value < 0.001) in the control group.
Publication Bias
The publication bias of the studies was also evaluated by regression, Egger test.
No significant publication bias was present in the reoperation rate (t = 1.50, p-value = 0.2072), mortality rate (t = 2.01, p-value = 0.1146), SAE rate (t = 1.24, p-value = 0.2841), and TCs rate (t = 2.29, p-value = 0.0619). However, the recurrence rate had a significant publication bias
(t = 2.67, p-value = 0.0282), which means the publication bias affected the recurrence rates in
the present study.
Discussion
The outcomes evaluated in this study have two main results: adhering to the primary
outcomes, the hematoma recurrence rate is lower in the TXA group than in the control
group. Additionally, TXA has not affected the mortality rate or TCs.
In our meta-analysis, the hematoma recurrence rate after surgery is 8%; however, in
the previous meta-analysis on the only RCTs and propensity score-matched studies performed
by Albalkhi et al found the rate of recurrence to be approximately 13.6%.[25] It should be noted that the risk of hematoma recurrence after drilling surgery varies
from 3 to 39%.[26]
[27]
[28] Our study shows a lower rate of hematoma recurrence than the Japanese studies with
60,000 participants, which found a rate of 13.1%.[29] The results of the present study attributed the high effect of Shibahashi et al,
which included 13,128 patients in their observational study, scored match study of
149,543 patients with 1-year follow-up in 1,100 hospitals. In the score-matched retrospective
study, the rate of hematoma recurrence that needed reoperation was significantly reduced
(TXA: 1.9% vs. surgery group: 6.1%, p-value < 0.001).[23] The major concern is that reoperation after hematoma recurrence is recommended for
symptomatic patients or radiological findings of cerebral compression.[30] Moreover, controversial results exist on the bias risk of different techniques in
the burr hole surgery.[3]
[31] So, findings should be consistent after sensitivity analysis.[23] In another large population cohort study, Miyakoshi et al included 465 patients
in each group of their research after the matching of 6,647 patients. They found a
reduction of relative risk of hematoma recurrence in the TXA versus control group
(6.5% vs. 16.8%) in the 6 months of follow-up.[22] Furthermore, the retrospective study by Tanweer et al on the role of TXA after surgical
evacuation found a 91.3% reduction in residual hematoma volume in patients who received
oral TXA, with no hematoma recurrence in the 3-month follow-up.[11] On the other hand, the randomized trial study by Wan et al found no postoperative
recurrence of hematoma in the TXA group. However, the was not large enough and 30%
of participants were lost to follow-up.[19] The discrepancy in results may be associated with the duration of follow-up, which
varies from 1 to 12 months.
The routine follow-up computed tomography (CT) scan is contemporary, while it cannot
predispose any other comorbidities related to the mortality,[32] and the hematoma recurrence is mainly affected by clinical symptomatic recurrence
following TXA or surgery alone.
In this study, no factors were associated with hematoma recurrence. However, several
factors have been demonstrated to be associated with the likelihood risk of hematoma
recurrence, including postoperative residual hematoma volume, the use of anticoagulants,
greater preoperative midline shift, and lack of intraoperative irrigation.[33]
[34] Zhu et al found male gender, bilateral hematoma, and drainage as significant predictors
of hematoma recurrence.[35] Also, Hamou et al[36] found expansion of hematoma as a predictor of hematoma recurrence. The anatomical
variations unifying system, for example, the Oslo CSDH Grading System, which includes
post- and preoperative hematoma volume and density changes in the CT scan,[37] could be considered to predict the hematoma recurrence.[38]
In our meta-analysis, the SDH volume reduction was only assessed in two studies[20]
[21] in which significant volume depletion was found. Gao et al's systematic review and
meta-analysis on 3,102 patients with traumatic brain injuries found that TXA significantly
reduced the hematoma volume.[39] Moreover, Yan et al, in their meta-analysis on the efficacy of TXA, found a reduction
of hematoma expansion of intracranial hemorrhage compared with the placebo.[40] However, another study on the efficacy of TXA in nontraumatic intracranial bleeding
found that TXA does not affect the severity of bleeding of subarachnoid hemorrhage
and intracranial hemorrhage.[41] It should be noted that the use of TXA in complex spine and pelvic surgeries was
effective in reducing blood loss in the case group rather than the control group.[42]
[43]
The secondary endpoint in our study was to examine if the administration of TXA as
an adjunct therapy increases the risk of thrombotic events in patients with CSDH.
Theoretically, TXA can promote thromboembolic events; however, previous large trials
have found that at doses 1 to 2 g, this adverse effect was not clinically significant.[44]
[45] In our analysis, four studies reported thrombosis in patients receiving TXA as an
adjunct treatment.[17]
[20]
[21]
[22]
[23] Our analysis found no association between TXA and the TC rate. Murao et al conducted
a meta-analysis on the effects of TXA on thrombotic events in bleeding patients and
found no evidence linking TXA to an increased risk of thrombotic events.[46] Another meta-analysis by Hariharan et al concluded that TXA was not associated with
an increased risk of thrombotic events.[47]
Although cranial surgery increases the 6-month mortality of CSDH,[48] our study showed that TXA after burr hole surgery has not affected the mortality
rate and SAEs of CSDH.[17]
[18]
[22] Furthermore, administration of TXA in traumatic brain injury did not affect the
mortality rate.[39] However, in the emergency settings, TXA reduced the 1-month mortality.[49] Future studies will be prompted to unify the scoring system of mortality and adverse
events in the setting of CSDH management. For instance, Zaki et al found beneficial
outcome prediction of mortality frailty index (mFI-5) in CSDH patients who underwent
cranial surgery.[50] Sastry et al also found mFI-5 as a prediction model of 90 days mortality in atraumatic
CSDH.[51]
Compared with other potential treatments, dexamethasone increased the adverse outcomes
following 6 months.[8] Additionally, in the systematic review and meta-analysis by Agrawal et al on 653
patients who received primary dexamethasone, 356 patients underwent surgery with an
odds ratio of 7.16%.[7] Furthermore, a previous meta-analysis conducted by Wang et al on 611 patients found
that atorvastatin was an effective option to reduce hematoma volume[9]; however, the inhibition of cytochrome P450[52] in high doses raises a concern about potential interactions. The results of our
study confirmed previous findings on the safety profile and known TXA interaction
with other drugs as the higher risk of polypharmacy in elderly patients with CSDH.[1]
[53]
Limitations
Although the results of our study showed promising results for the role of TXA as
an adjunctive therapy to the surgical evacuation of CSDH, the variation of follow-up
time in the studies and the number of patients lost in the clinical course is the
notable limitation of this study. Additionally, the long-term outcomes of TXA on the
hematoma recurrence have not been described in the studies included in our analysis.
Finally, the study's results may be affected by factors such as the hematoma pattern,
the history of CSDH, and existing comorbidities. Relying on these important covariation
factors and inadequate reporting of the outcomes related to these factors, subgroup
analysis was not feasible enough. So, future studies should note the role of comorbidities
that influence the outcome of surgery for CSDH patients.
Conclusion
This study indicated that TXA as an adjunct treatment to BC is a safe option to reduce
postoperative hematoma recurrence and reoperation, and it has not increased the rate
of TCs and SAEs.
