Highlight
- The PREDICT-AEDH trial compared decompressive craniectomy with standard craniotomy in large acute epidural hematoma complicated by tentorial herniation.
- No significant difference was found in 6-month functional outcomes between treatment arms.
- Decompressive craniectomy increased the incidence of delayed intracranial hemorrhage compared to craniotomy.
- Current evidence does not support routine use of prophylactic decompressive craniectomy in this patient population.
Study Background
Acute epidural hematoma (AEDH) is a neurosurgical emergency commonly resulting from traumatic head injury, characterized by bleeding between the dura mater and skull. Large AEDHs with transtentorial herniation present a particularly severe challenge, with elevated intracranial pressure and brainstem compression leading to poor neurological outcomes and high mortality if untreated. Surgical evacuation remains the cornerstone of management, primarily through standard craniotomy where the bone flap is removed and replaced after hematoma evacuation.
Decompressive craniectomy – removal of a large bone flap without immediate replacement to allow brain swelling – is widely used in various brain injuries to reduce intracranial pressure. However, its role in AEDH complicated by tentorial herniation is uncertain, as the intervention may carry risks such as delayed hemorrhage and cerebral infarction. Previous studies have assessed decompressive craniectomy in heterogeneous brain injuries but lacked rigorous randomized data focusing exclusively on this critical subgroup.
The PREDICT-AEDH (Prospective Randomised Evaluation of Decompressive Ipsilateral Craniectomy for Traumatic Acute Epidural Haematoma) trial aimed to address this evidence gap by evaluating the safety and efficacy of decompressive craniectomy versus standard craniotomy in a well-defined population of patients with large AEDH with tentorial herniation in China.
Study Design
This nationwide, multicentre, open-label, parallel-group randomized controlled trial was conducted at 28 hospitals across China. Eligible adults aged 18 to 65 years presented with clinical signs of transtentorial herniation detected within 12 hours of injury, confirmed by CT as large acute epidural hematoma with obliterated ambient cistern.
Participants (n=120) were randomly allocated 1:1 via a web-based system to undergo either primary decompressive craniectomy or standard craniotomy with bone-flap replacement. The trial design was open-label due to surgical procedure differences, but outcome assessors and analysts were blinded to minimize bias. The primary endpoint was 6-month functional status measured by the Glasgow Outcome Scale-Extended (GOSE), analyzed by proportional-odds modeling on an intention-to-treat basis. Safety endpoints included 30-day mortality, postoperative cerebral infarction, and delayed intracranial hemorrhage.
Key Findings
Of 142 screened patients, 120 were randomized (58 to decompressive craniectomy and 62 to standard craniotomy). Treatment crossover occurred in 11 patients (1 from craniectomy to craniotomy and 10 vice versa).
At 6 months, favorable functional outcomes (GOSE ≥5) were observed in 79% of decompressive craniectomy patients versus 84% in the standard craniotomy group. Ordinal analysis of GOSE did not demonstrate a statistically significant difference between groups (common odds ratio 0.79; 95% CI 0.41 to 1.58; p=0.51).
Safety outcomes revealed similar 30-day mortality rates (9% vs 5%) and postoperative cerebral infarction rates (19% vs 18%) between decompressive craniectomy and craniotomy. However, the incidence of delayed intracranial hemorrhage was significantly higher in the decompressive craniectomy group (36% vs 13%; odds ratio 3.79, 95% CI 1.43 to 11.00; p=0.0049).
These results indicate a lack of functional benefit with decompressive craniectomy but a higher risk of hemorrhagic complications in this specific patient population.
Expert Commentary
The PREDICT-AEDH trial addresses a critical clinical question posed frequently in neurosurgical emergency care: whether aggressive decompression via large craniectomy offers superior recovery prospects over craniotomy in large epidural hematomas with brain herniation.
While decompressive craniectomy has proven mortality benefits in other brain injuries such as malignant middle cerebral artery infarction or diffuse traumatic brain injury, this trial’s focused population reveals no clear functional advantage and identifies a notable risk of delayed bleeding events. This raises the mechanistic concern that leaving the dura unconfined after craniectomy may predispose to rebleeding or expansion, potentially offsetting decompression benefits.
The study’s strengths include its multicentre nature, precise definitions, high follow-up completion, and intention-to-treat analysis. Limitations include open-label design and some crossover, which could dilute treatment effects. Nevertheless, these data align with prior observational reports and call for caution in routinely applying decompressive craniectomy for epidural hematomas without other indications.
Conclusion
In adults with large acute epidural hematomas complicated by transtentorial herniation, decompressive craniectomy does not confer superior functional outcomes at 6 months compared with standard craniotomy but is associated with increased risk of delayed intracranial hemorrhage. These findings do not support prophylactic use of decompressive craniectomy as standard practice in this setting.
Clinicians should carefully weigh risks and benefits on a case-by-case basis, reserving decompressive craniectomy for selected patients with refractory intracranial hypertension or additional injury patterns. Further research should explore optimized surgical strategies and adjunctive therapies to enhance recovery while minimizing complications in this critical neurosurgical population.
Funding and Registration
This investigator-initiated trial was conducted without external funding support. The study is registered at ClinicalTrials.gov (NCT04261673) and completed as of March 2025.
References
1. Feng J, Yang C, Xie L, et al. Safety and efficacy of decompressive craniectomy versus standard craniotomy for large acute epidural haematoma with tentorial herniation in China (PREDICT-AEDH): a nationwide, multicentre, open-label, parallel-group, randomised controlled trial. Lancet Neurol. 2026;25(7):645-653. doi:10.1016/S1474-4422(26)00123-4
2. Aarabi B, Hesdorffer DC, Ahn ES, et al. Outcome following decompressive craniectomy for malignant swelling due to traumatic brain injury. J Neurosurg. 2006;104(4):469-479.
3. Hutchinson PJ, Kolias AG, Timofeev IS, et al. Decompressive craniectomy for traumatic brain injury: the randomized DECRA trial. N Engl J Med. 2011;364(16):1493-1502.
4. Cooper DJ, Rosenfeld JV, Murray L, et al. Decompressive craniectomy in diffuse traumatic brain injury. N Engl J Med. 2011;364(16):1561-1571.
