Safety and feasibility of transitioning from conventional burr-hole craniostomy to endoscope-assisted evacuation for chronic subdural hematoma: A retrospective cohort study

Introduction

Chronic subdural hematoma (cSDH) is a common neurosurgical disorder characterized by the gradual accumulation of blood and its degradation products within the subdural space. The annual incidence of cSDH is estimated to be approximately 13.5 cases per 100,000 individuals, with an almost fivefold increase among those older than 65 years. ¹ As the global population continues to age, the clinical and socioeconomic burden of the disease is steadily increasing.

Traditionally, the surgical management of cSDH has encompassed a range of procedures, from minimally invasive burr-hole craniostomy to formal craniotomy for cases involving complex anatomy or organized hematomas. ² Burr-hole craniostomy remains the most widely adopted technique because of its simplicity and favorable safety profile; however, postoperative recurrence and residual septated hematomas remain persistent challenges.

In recent years, endoscopic techniques have been increasingly incorporated into the surgical treatment of cSDH. Endoscope-assisted surgery provides enhanced visualization of the subdural cavity, enabling precise identification and fenestration of internal septations while minimizing manipulation of the brain parenchyma. Accumulating evidence suggests that endoscopic assistance may reduce postoperative recurrence rates and procedure-related complications compared with conventional evacuation methods.^1,3–5

Nevertheless, the adoption of novel neurosurgical techniques requires a transition from established surgical paradigms to innovative approaches. This transition is often accompanied by a learning curve, which may initially result in prolonged operative times and potentially influence perioperative outcomes. Despite the theoretical advantages of endoscopic assistance, the practical challenges encountered during the early phase of implementation—and their impact on patient prognosis—have not been fully elucidated.

In the present study, we conducted a retrospective comparative analysis of conventional burr-hole craniostomy and endoscope-assisted surgery for cSDH at Far Eastern Memorial Hospital. We aimed to determine whether the integration of endoscopic assistance leads to improved clinical outcomes and patient prognosis compared with the conventional approach. Additionally, we evaluated the technical challenges associated with the adoption of endoscopic surgery and characterized the learning curve during its implementation. By addressing these factors, this study seeks to provide valuable evidence to inform surgical decision-making and optimize future management strategies for patients with cSDH.

Method

Result

A total of 110 patients were included in our study, with 60 patients in the treatment group and 50 in the control group (Table 1). The treatment group was significantly older than the control group (74 ± 11.54 vs. 67.56 ± 12.32 years; p = 0.006), though no significant differences were observed in gender (p = 0.35) or hematoma location (p = 0.436). Clinical presentations were similar across both groups (p = 0.83), with gait disturbance being the most frequent symptom (73% vs. 74%), followed by cognitive impairment (10% in each group), headache (8.3% vs. 14%), seizures (5% vs. 2%), and aphasia (5% vs. 6%). Baseline functional status and severity, as measured by pre-operative GCS (p = 0.93) and mRS scores (p = 0.65), as well as the history of antithrombotic therapy (p = 0.68), were comparable between the two groups. Regarding surgical and functional outcomes, 90.0% of the treatment group and 89.36% of the control group returned to their pre-symptomatic baseline (representing the state prior to cSDH-related neurological impairment) within a 3-month follow-up period. The treatment group required a significantly longer median operation time than the control group (120.0 min [IQR 90.0–150.0] vs. 90.0 min [IQR 76.3–127.5]; p = 0.003). Despite the longer operative duration, there were no statistically significant differences in clinical outcomes at discharge, including GCS (p = 0.29), mRS (p = 0.08), complication rates (5% vs. 6%; p = 1.00), or recurrence rates (1.69% vs. 4.17%; p = 0.586). Additionally, the time to drain removal (2.73 ± 1.04 vs 2.66 ± 0.15 days; p = 0.72) and hospital stay duration (7.48 ± 5.21 vs 7.34 ± 8.42 days; p = 0.91) remained consistent across both groups.

OPEN IN VIEWER

Table 1.

Clinical characteristics and outcome comparison for chronic subdural hematoma (cSDH) between treatment and control group.

Characteristics	Treatment	Control	p-value
Number of patients	60	50	​
Gender (M/F)	47/13	42/8	0.35
Age (year)	​		0.006
mean ± SD	74 ± 11.54	67.56 ± 12.32	​
range	26 - 91	39 - 87	​
Location (R/L/B)	21/27/12	17/18/15	0.436
Clinical symptom – no. (%)	​		0.83
Gait disturbance	44 (73%)	37 (74%)	​
Cognitive impairment	6 (10%)	5 (10%)	​
Headache	5 (8.3%)	7 (14%)	​
Seizure	3 (5%)	1 (2%)	​
Aphasia	3 (5%)	3 (6%)	​
Antithrombotic therapy– no. (%)	​		0.68
Anticoagulant	4 (6.7%)	4 (8%)	​
Antiplatelet	12 (20%)	8 (16%)	​
GCS (mean ± SD)
Before operation	14.28 ± 1.35	14.26 ± 1.45	0.93
When discharge	14.72 ± 1.6	14.96 ± 0.2	0.29
mRS (mean ± SD)
Before operation	3.52 ± 0.85	3.44 ± 0.91	0.65
When discharge	2.1 ± 0.97	1.74 ± 1.16	0.08
Operation times (minutes)	​		0.003
median (IQR)	120 (90 – 150)	90 (76.25 – 127.5)	​
			​
Complications (%)	5	6	1.00
Recurrent (%)	1.69	4.17	0.586
Time to drain removal (days)	​	​	0.72
mean ± SD	2.73 ± 1.04	2.66 ± 0.15	​
range	1 - 5	1 - 4	​
Hospital stays (days)	​	​	0.91
mean ± SD	7.48 ± 5.21	7.34 ± 8.42	​
range	2 - 37	3 - 47	​
Recover to pre-symptomatic baseline (%)	90	89.36	1.00

To evaluate the independent impact of the surgical approach on clinical outcomes, a multivariable logistic regression analysis was performed (Table 2). After adjusting for potential confounders, including operative time, age, antithrombotic therapy, and pre-operative mRS, the surgical method was not significantly associated with the likelihood of recovery to the pre-symptomatic baseline (aOR: 1.28; 95% CI: 0.33–4.95; p = 0.725). Notably, longer operative duration did not adversely affect functional recovery (aOR: 1.00; p = 0.841), suggesting that the technical demands of the endoscopic approach do not compromise clinical efficacy.

OPEN IN VIEWER

Table 2.

Univariate and multivariable logistic regression analysis of factors associated with recovery to functional baseline.

Variable	Univariate OR (95% CI)	p-value	Multivariable aOR (95% CI)	p-value
Surgical Method (Endo vs. BH)	1.07 (0.31–3.75)	0.914	1.28 (0.33–4.95)	0.725
Operative Time (minutes)	1.00 (0.99–1.02)	0.817	1.00 (0.99–1.02)	0.841
Age (years)	0.96 (0.90–1.02)	0.204	0.97 (0.90–1.04)	0.397
Antithrombotic Therapy	0.58 (0.16–2.17)	0.421	0.68 (0.17–2.67)	0.583
Pre-operative mRS	0.54 (0.21–1.39)	0.200	0.60 (0.22–1.65)	0.325

A total of six complications were identified, distributed equally between the treatment and control groups (n = 3 each; Table 3). In the treatment group, complications included one re-operation due to post-operative acute subdural hematoma, one conversion to craniotomy due to intractable intraoperative bleeding, and one mortality. Notably, only the patient who underwent conversion to craniotomy successfully achieved recovery to the pre-symptomatic baseline at the 3-month follow-up. In the control group, complications consisted of one surgical site infection (occurring 3 weeks post-operatively and requiring a 3-week course of antibiotics), one case of cerebral vasospasm (presenting as consciousness deterioration one day post-operation, with CT angiography confirming long-segment narrowing of the left distal internal carotid artery and the M1 segment of the middle cerebral artery), and one mortality. Within this group, only the patient with the wound infection recovered to their pre-symptomatic baseline. Of particular note, the mortality case in the control group was the only patient among all complicated cases who was receiving antithrombotic therapy. Importantly, neither of the mortalities was directly related to surgical complications; both deaths were attributed to sepsis secondary to hospital-acquired pneumonia.

OPEN IN VIEWER

Table 3.

Clinical characteristics of the complications.

Case	Gender	Age	Endoscopy	Location (R/L/B)	Complication	Antithrombotic therapy	Recover to base line
1	M	65	No	L	Wound infection	No	Yes
2	F	44	No	R	Vasospasm	No	No
3	M	85	No	B	Mortality	Antiplatelet	No
4	M	65	Yes	B	Re-operation	No	No
5	M	75	Yes	R	Convert to craniotomy	No	Yes
6	F	90	Yes	R	Mortality	No	No

To assess the impact of individual surgical experience on procedural efficiency, we performed a subgroup analysis of operative durations across different surgeons performing the endoscopic procedure. To ensure consistency and minimize confounding variables, patients with bilateral cSDH were excluded; thus, the analysis focused exclusively on unilateral procedures. A total of 48 patients underwent unilateral endoscopic evacuation, with an overall mean operative time of 113.96 ± 39.50 minutes.

Figure 1 records the operative time trends for the four surgeons who performed at least five unilateral endoscopic-assisted procedures. By plotting the case sequence against the operative duration, these curves demonstrate the transition in surgical efficiency as each surgeon gained cumulative experience with the endoscopic technique.OPEN IN VIEWER

Discussion

The therapeutic approach to cSDH is primarily dictated by the patient’s clinical status. For individuals presenting without neurological symptoms, conservative management is increasingly centered on statin therapy as the primary intervention, alongside intracranial pressure control and serial monitoring. ⁷ Consistent with this approach, statin therapy was initially administered to only one patient in our treatment group who presented with cSDH but no initial neurological deficits. This strategy aligns with the paradigm shift highlighted in a multicenter registry study by Tao Liu et al., where atorvastatin is utilized for its anti-inflammatory properties to promote hematoma absorption and enhance recovery in patients who do not require immediate decompression. ⁸ This patient received Atorvastatin (20mg daily) for three months; nonetheless, the cSDH failed to resolve radiographically, and clinical symptoms—specifically a persistent headache—eventually worsened. This clinical progression necessitated a transition to surgical intervention. Our experience underscores that while statin therapy is a viable first-line option for asymptomatic patients, surgical evacuation remains the definitive treatment when medical management proves ineffective or when clinical deterioration occurs. ⁹

The recommendation regarding early mobilization, particularly the findings from the GET-UP Trial, is highly valued. We acknowledge that current evidence suggests early mobilization can significantly reduce systemic medical complications (such as pneumonia and deep vein thrombosis) without necessarily increasing the recurrence rate of cSDH. ¹⁰ However, our institutional protocol currently favors maintaining a flat supine position until drain removal based on the following clinical considerations. First, prevention of accidental dislodgement: In our clinical experience, early mobilization with an indwelling subdural drain—especially in the elderly population—poses a significant risk of accidental drain dislodgement, which may lead to the need for surgical revision. Second, fall prevention: Many cSDH patients present with preoperative hemiparesis or gait instability. We have observed that mobilization while the drain is still in situ can increase the risk of falls due to limb weakness combined with the physical encumbrance of the drainage system. Third, local hemodynamics: While head-on-bed elevation may not significantly affect recurrence, we prioritize the supine position to facilitate brain re-expansion and maximize the siphoning effect of the drain during the critical first 24–48 hours. ¹¹ Ultimately, while we acknowledge the systemic benefits of early mobilization, our protocol prioritizes patient safety and catheter integrity during the immediate postoperative phase. To balance these competing concerns, our protocol mandates early mobilization as soon as the subdural drain is removed. This ensures both postoperative safety and the functional benefits of early physical activity.

Effective evacuation of cSDH is well-established to improve patient outcomes and mitigate the risk of recurrence. ¹² Current literature indicates that morbidity and mortality rates for standard craniostomy remain low, ranging from 4.0–9.3% and 2.5–3.7%, respectively, with reported recurrence rates between 10.5% and 12.0%.^1,13–15 Recent institutional data from our center demonstrate that our clinical outcomes regarding morbidity, mortality, and recurrence following craniostomy are now closely aligned with these international benchmarks. Consequently, our surgical team has begun exploring advanced procedural alternatives to further optimize patient recovery and clinical prognosis.

The clinical application of endoscopic surgery for cSDH management dates back to 1988, as pioneered by Karakhan. ¹⁶ Since this seminal description, a growing body of literature has further validated the safety and efficacy of endoscopic techniques in treating this condition.^5,17,18 Endoscopy provides enhanced intraoperative visualization during evacuation, allowing for the clear identification of trabeculae and septations even when utilizing a burr-hole approach. This superior visual field facilitates a more exhaustive hematoma evacuation, precise excision of neomembranes, and meticulous microscopic hemostasis. ¹⁹

Our study demonstrates that transitioning from conventional burr-hole craniostomy to endoscopic evacuation for cSDH is both safe and clinically feasible. While the treatment group was significantly older than the control group (p = 0.006), our data confirm that this chronological age gap did not translate into a difference in baseline functional status, as measured by the pre-operative mRS score (p > 0.05). Although a patient’s general medical history provides a broad view of patient health, we focused on antithrombotic therapy as the primary clinical surrogate for baseline risk. For cSDH patients, particularly the elderly, the management of antiplatelet and anticoagulant medications represents a more significant surgical challenge than many chronic systemic diseases. By analyzing this factor alongside pre-operative mRS scores, we believe we have captured the most relevant baseline parameters that influence surgical outcomes and functional recovery in this cohort. This age discrepancy likely reflects a patient-led selection bias; elderly patients and their families often perceive the endoscopic technique as a “minimally invasive” and inherently safer alternative, which likely drove higher acceptance among the older cohort. While our cohorts exhibited a significant difference in chronological age, our multivariable-adjusted model demonstrated that age was not an independent driver of the observed functional outcomes in this study. Furthermore, we provide a more transparent assessment of the precision of our surgical outcomes, addressing the inherent variability in geriatric cSDH populations. The temporal clustering of the treatment group starting in Q4 2023 (Figure 2) underscores a shift in surgical adoption. The endoscopic approach yielded clinical outcomes—including GCS and mRS scores at discharge—that were comparable to conventional methods. Specifically, 90% of patients in the treatment group successfully returned to their pre-symptomatic baseline within three months, a rate nearly identical to that of the younger control group (89.36%). These findings suggest that the endoscopic technique is particularly well-suited for the geriatric population, who often prioritize safety and reduced surgical trauma. ²⁰ OPEN IN VIEWER

Figure 1 illustrates the learning curves of the four surgeons, where an initial phase of longer operative durations is evident. As surgeons gained proficiency and adjusted to the endoscopic technique through consecutive cases, a progressive reduction in operative time was observed. While the treatment group’s overall average was significantly higher than the control group (123.75 ± 45.21 vs. 104.3 ± 41.97 minutes; p = 0.02), this trend suggests that the initial technical demands of endoscopic visualization are overcome with experience. This extended operative time did not translate into higher morbidity. The treatment group showed a lower recurrence rate compared to the control group (1.69% vs. 4.17%), though this difference did not reach statistical significance.

Safety analysis further supports the transition, as complication rates were nearly identical between groups (5% vs. 6%). A detailed review of complications in the treatment group revealed specific challenges inherent to the learning phase, such as one case requiring re-operation and another requiring conversion to craniotomy. Nevertheless, the overall stability of hospital stay durations and recovery outcomes confirms that the endoscopic approach provides a reliable alternative to conventional surgery without compromising patient safety during the institutional transition period.

Endoscopic intervention is essential for confirming the complete evacuation of residual hematomas under direct visualization, particularly in cases of cSDH characterized by internal septations. Conventional burr-hole craniostomy often fails to adequately debride these organized fibrous membranes and sequestered blood, which subsequently restricts cerebral re-expansion and inhibits the obliteration of the subdural space. In contrast, endoscopy facilitates comprehensive hematoma clearance and the fenestration of internal septa. A meta-analysis by Songyi Guo et al. demonstrated that endoscope-assisted surgery significantly reduces both recurrence and postoperative complication rates compared to conventional methods. ³ Furthermore, Toshiyuki Amano et al. reported on 97 patients treated between 2013 and 2019, achieving a 0% reoperation rate for re-bleeding and 0% mortality, with a mean operative time of 59.4 ± 19.2 minutes; they concluded that endoscopy is superior for managing complex hematomas involving organized clots or fibrous septa. ⁵

A notable limitation of the present study is the lack of routine postoperative imaging to objectively quantify residual hematoma volume. In our clinical practice, we recognize that the brain does not immediately re-expand to its original position following evacuation of a cSDH due to prior compression. Therefore, routine postoperative brain CT is not performed for all patients at our institution. Residual subdural space on imaging, in the absence of corresponding clinical symptoms, is a common finding and does not necessarily warrant further intervention.^21,22 Accordingly, postoperative evaluation in our study was primarily based on clinical recovery rather than routine imaging findings, and we did not systematically quantify residual hematoma volumes using postoperative CT scans. While this approach reflects real-world clinical practice, it may limit the ability to directly demonstrate the incremental value of endoscopic visualization in detecting residual collections. Future studies incorporating standardized imaging protocols are warranted to further elucidate the potential advantages of endoscopic techniques in identifying and managing residual hematomas.

Based on our clinical experience, there remains a substantial opportunity to optimize operative efficiency and further reduce reoperation rates. We anticipate that, as institutional volume and surgical proficiency continue to expand, procedural times will decrease and clinical outcomes will improve, potentially enhancing overall patient prognosis. This hypothesis warrants validation through future large-scale, prospective, multicenter studies.

Despite its advantages, endoscopic surgery for cSDH presents several inherent challenges and disadvantages that must be meticulously managed. The primary technical drawbacks include a cumbersome equipment setup and the demanding coordination required to manipulate the camera and suction simultaneously, which in certain instances necessitates specialized tools, such as a flexible suction cannula. From a safety perspective, the procedure carries a risk of cortical surface damage resulting from blind angles, surgical errors, or the stripping of membranes that are densely adherent to the cortex. Furthermore, the technical complexity of integrating endoscopic visualization typically results in an extended operation time.^4,17,19

Conclusion

For cSDH presenting with neurological deficits, both conventional burr-hole craniostomy and endoscope-assisted evacuation are safe and effective therapeutic approaches. Transition from traditional techniques to an endoscopic approach does not increase patient complication rates or compromise clinical outcomes; even during the learning curve characterized by longer operative durations and an older patient demographic, the endoscopic group achieved functional recovery comparable to conventional methods. We believe that through the accumulation of surgical experience, patients can derive maximal benefit from this technique, which is both learnable and worthy of widespread promotion as a reliable alternative to conventional surgery.