Balancing Bleeding Risk and Thromboembolic Complications in Elderly Chronic Subdural Hematoma Patients Undergoing Burr Hole Trephination : A Multicenter Retrospective Cohort Study and Literature Review

Statistical analysis

Data acquired included the median patient age, the length of time between discontinuation of AT and surgery, and the length the postoperative period requiring drain maintenance. First and third quartile were confirmed using interquartile range. The chi square test was utilized for assessment of a correlation between hematoma density and pattern on the initial brain CT and for the presence of acute hemorrhage on postoperative CT with antiplatelet agent usages. The correlations between initial hematoma density and pattern and postoperative acute hemorrhage with anticoagulant usage were assessed by Fisher’s exact test. The correlations between recurrence and reoperation and medication history of AT were confirmed by chi-square and Fisher exact tests. Generalized logistic regression was used for analyzing the period from hospitalization to surgery and the length of drain usage. R studio (version 4.1.2; R Foundation for Statistical Computing, Vienna, Austria) was used.

Baseline characteristics

The median age of the patients was 75 years (67.00–81.00 years), and 311 males (67.32%) and 151 females (32.68%) were included in the study. Hypertension was present in 262 patients (56.71%), 140 patients (30.30%) had diabetes, and 84 patients (18.18%) had hyperlipidemia. Liver disease affected 28 patients (6.06%), and 24 patients had renal disease (5.19%). A history of ischemic cerebral infarction was confirmed in 37 patients (8.01%), hemorrhagic stroke in three patients (0.06%), and both ischemic and hemorrhagic stroke in one patient (0.22%). A history of heart disease was present in 50 patients (10.82%), cancer in 31 patients (6.71%), and blood disease in 10 patients (2.16%). For CSDH, right side operation was performed in 183 cases, left side operation was performed in 201 cases, and bilateral operation was performed in 78 cases (Table 1).

Table 1.

Baseline characteristics

Details on history of AT administration

Regarding medication history, 100 patients were using AP, 50 for secondary prevention of heart disease, and 37 for secondary prevention of ischemic stroke. The other patients took AP for primary prevention of stroke depending on medical histories of issues such as hypertension or diabetes mellitus. Seventeen patients were using AC, and four for secondary prevention of ischemic stroke and the others for secondary prevention of heart disease. Two patients were using both an AP and an AC together (AP+AC) because of history of both heart disease and ischemic stroke. Details regarding the types of AT administration are outlined in Table 2.

Table 2.

Medication history of antiplatelet agents and anticoagulants

Timing of surgery

Patients with no medication history of AT required a median of 1 day (0.00–2.00) of hospitalization before surgery, and the same was true for patients taking AP (0.00–3.00). A median of 2 days (1.00–6.00) of presurgical hospitalization, for patients taking AC, was the longest interval of pre-surgical hospitalization. Of the two patients using AP+AC, one underwent surgery after 2 days and the other after 12 days (Fig. 1). Use of AP did not significantly delay BHT (p=0.318), but use of AC did significantly delay the operation (p=0.048). In patients taking AC, further analysis was performed depending on type. Being prescribed warfarin was of borderline significance (p=0.076) for a longer surgical delay, but there was no clear correlation between factor Xa inhibitor or thrombin inhibitor prescription (p=0.340 and p=0.558) and hospital stay prior to surgery.

Fig. 1.

The interval between discontinuation of medication and surgery by medication group. AP : antiplatelet agents, AC : anticoagulants.

Duration of in-dwelling drainage catheter

The duration of postoperative subdural drainage via indwelling catheter was a median of 2 days (1.00–3.00) in patients with no medication history of AT, a median of 2 days (1.00–3.00) in the AP group, and a median of 3 days (2.00–3.00) in the AC group. In the association analysis, even though there was no difference in median, AP intake significantly increased the drain period (p=0.026), as did AC intake (p=0.037) (Fig. 2).

Fig. 2.

Duration of postoperative in-dwelling drainage catheter by medication group. AP : antiplatelet agents, AC : anticoagulants.

Postoperative acute hemorrhage

On brain CT performed the first day after surgery, acute hemorrhage was confirmed in 21.70% of the drug-free group, 19.00% of the AP group, 35.29% of the AC group, and all patients taking both AP+AC. AP intake had no effect (p=0.808), but AC intake was significantly associated with acute hemorrhage (p=0.044) (Table 3). We performed a subgroup analysis of patients taking AP. The period of AP discontinuation prior to surgery was not statistically significant with the incidence of acute hemorrhage on postoperative (p=0.161).

Table 3.

Clinical characteristics according to medication history of antiplatelet agents and anticoagulant

Recurrences and reoperations

Recurrence occurred in 17.78% of cases of the no medication group, and reoperation was performed in 14.29%. Reoperation was performed in all cases of recurrence in the AP and AC groups, and the rates were 12.00% and 5.88%, respectively. Recurrence occurred in both patients taking combined AP+AC, and reoperation was performed in one of them. Regarding the association with recurrence, intake history of AP or AC had no significant effect (p=0.517 and p=1.000, respectively), and there was no effect on reoperation rate (p=0.924 and p=1.000) (Table 3). Additionally, a statistically significant positive correlation was observed between the preoperative AP discontinuation period and the recurrence rate (p=0.033; odds ratio, 1.658; 95% confidential interval, 1.036–2.673).

Morbidity and mortality

Morbidity occurred in 4.37% of patients with no medication history and included cerebral contusion, acute subdural hemorrhage, pneumonia, and abscess formation after BHT. Morbidity occurred in 5.00% of patients taking AP and included epidural hemorrhage, subarachnoid hemorrhage, and acute kidney injury after the operation. Morbidity occurred in one patient taking AC as acute subdural hemorrhage. There was also one case of morbidity as acute subdural hemorrhage in patients taking AP+AC. There was no statistical correlation between medication history and morbidity for AP or AC (p=0.763 and p=0.800, respectively). Mortality was confirmed in 2.04% of the no medication group, 3.00% of the AP group, and 0% of the AC and AP+AC groups. There was no significant correlation between mortality and intake of AP or AC (p=0.748 and p=0.992, respectively) (Table 3).

Use of AT and recurrence of CSDH

There are several retrospective studies that have analyzed the relationship between AT, especially AP, and the postoperative course of CSDH. Most of these studies concluded that AT use did not affect recurrence of CSDH after the operation [1,5,7,15,25]. One report showed a relationship between AP and recurrence of CSDH, but the recurrence rate in the article was relatively higher than in other articles, suggesting a bias [21]. Another study provided evidence that medication history of AC affected recurrence of CSDH; but, in the AT group, a larger preoperative hematoma volume was noted, with no effect of AP on recurrence [27]. In separate study, it was observed that while AT usage did not have a significant impact on recurrence in univariate analysis, an increase in odds ratio was identified in univariate regression analysis [5]. Like the results of our study (p=0.517, Table 3), the effect of AT usage on recurrence of CSDH is not clear (Table 4). Additionally, preoperative AP discontinuation period and recurrence rate showed a positive correlation in this study. This may reflect the paradoxical risk of discontinuation of AT previously revealed [2].

Table 4.

Review of previous articles concerning the relationship between AT and clinical course of postoperative CSDH patients

Management of AT : before BHT (whether and when to stop AT)

In existing literature, there are varying opinions on whether and when to stop antiplatelet medication before surgery for CSDH. In a retrospective study of patients taking AT, discontinuation of medication did not affect the clinical outcome [19]. Considering the characteristics of platelets, it is believed that the effectiveness of the medications would diminish when AP is discontinued for 5 days. However, a review article suggests that AP should be stopped 1 week before surgery, as with other major surgeries [17]. On the other hand, there are studies suggesting that even when discontinued for more than 3 days, it is possible to reduce the occurrence of postoperative acute hemorrhage [21,25]. In our study, there was no significant correlation between the duration of AT discontinuation before surgery and the occurrence of acute hemorrhage after surgery (Fig. 2).

Management of AT : after BHT (When to restart AT)

A separate study found that early resumption group had a higher recurrence rate [15]. However, there are also important retrospective studies concerning thromboembolic complications after CSDH surgery in patients with history of AT. Most studies emphasize the importance of early resumption of AT after surgical intervention for CSDH [2,7,10,26]. Resumption timing of AT after operation did not affect recurrence of CSDH in two studies [7,26], and patients with history of AT use showed more frequent thromboembolic events after the operation in three studies [2,10,26] (Table 4). In our dataset, it was observed that AT was resumed in 25 cases, with a wide range of timeframes spanning from 2 days to 3 months. However, due to limited number of patients and the lack of detailed records regarding the exact timing of resumption, we were unable to perform an analysis on this aspect.

Paradigm shift to continuous use of AT for CSDH patients undergoing BHT

The importance of continuous use of AT is an emerging concept in our aging society. Regarding the risk of bleeding during surgery, studies on dosing and discontinuation guidelines for AP and AC for cardiac surgery and non-cardiac surgery are on-going; and guidelines are currently in place [18]. Major adverse cardiovascular event is a complication of surgery using general anesthesia and occurs in close to 5% of patients over the age of 45 years [25]. For patients aged 40 to 70 years at high risk for CVD, aspirin is also recommended for primary prevention [3]. In addition, lifelong AP prescription is recommended for secondary prevention in patients with CVD [14] . A more recent large observational study reported that maintenance of aspirin did not increase the risk of bleeding in noncardiac surgery [16]. Based on this, patients with CVD taking aspirin are recommended to maintain aspirin therapy alone when undergoing non-cardiac surgery [9,18]. The common belief is that maintaining medications, especially aspirin, is more helpful than discontinuation. Also, treating patients more actively rather than delaying surgery out of fear of recurrence of CVD and discontinuing medication is preferable.

In our study, patients who took AP or AC had longer postoperative maintenance of in-dwelling drains than those who did not take AT. Although more frequent acute bleeding was observed on postoperative CT in patients taking AC, there was no effect on recurrence or reoperation. Close observation and management of patients immediately after operation are essential for bleeding control. Although this study did not prove a clear difference between warfarin and other AC, direct oral anticoagulant (DOAC) therapy has a lower risk of bleeding than use of warfarin, a vitamin K antagonist; and DOAC is beneficial because there are fewer restrictions such as INR control [4]. The use of DOAC is increasing, and the risk of drug-induced bleeding in patients with CSDH is expected to decrease. Related studies are needed in the future.

Limitations of the study

In conducting a multi-center retrospective study, discontinuation of AT was performed at the discretion of individual surgeons. Also, there was the possibility of non-uniform data collection in the multicenter data collection process. Compared to AP, the analysis of AC was limited due to the small number of patients taking AC. Also, our dataset lacked information on resumption of AT and thromboembolic complications. To establish guidelines on the discontinuation and resumption of AT medications in CSDH, a larger prospective randomized study is necessary.