TL;DR
This study investigated whether decompressive craniectomy (DC) or craniotomy leads to better outcomes for patients with acute subdural hematomas (ASDH). Despite significant variation in surgical practice across European centers, the study found no difference in functional outcomes between the two procedures, suggesting that primary DC should be reserved for specific cases of severe brain swelling.
What Problem Does This Paper Address?
Traumatic acute subdural hematomas (ASDH) are serious injuries that can lead to brain herniation and death. Surgical evacuation is necessary, but there's been ongoing debate about the best approach: craniotomy (where the bone flap is replaced) or decompressive craniectomy (DC, where the bone flap is removed and replaced later). While DC can help manage intracranial pressure, it's more invasive and carries risks. Until recently, there was limited high-quality evidence to guide this decision, leading to significant variation in how different hospitals approached these cases. This study aimed to clarify the comparative effectiveness of these two surgical strategies in a real-world setting, outside of a controlled trial.
Methods
The researchers conducted a prospective, observational cohort study using data from the Collaborative European NeuroTrauma Effectiveness Research in TBI (CENTER-TBI) and NeuroTraumatology Quality Registry (Net-QuRe) studies. They employed an instrumental variable (IV) analysis, a powerful statistical technique for observational studies. The IV analysis uses the variation in treatment preference between centers as a proxy for treatment assignment. This helps to reduce confounding by indication, meaning it accounts for the fact that doctors might choose a particular surgery for patients who are already more or less likely to have a good outcome.
Specifically, they used the "case-mix adjusted proportion of DC per center" as their instrumental variable. This means they looked at how often each center performed DC, after accounting for patient characteristics. They then compared functional outcomes (measured by the Glasgow Outcome Scale Extended - GOSE) between centers with different preferences for DC versus craniotomy. They also analyzed secondary outcomes like in-hospital mortality, need for further surgeries, and quality of life.
Datasets
The study utilized data from the CENTER-TBI (Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury) and NeuroTraumatology Quality Registry (Net-QuRe) studies. These prospective, multicenter observational studies enrolled patients across Europe and Israel between 2014 and 2020. The specific cohort for this study included 336 patients who underwent acute neurosurgical evacuation for ASDH.
Key Insights and Ideas
- Practice Variation is Significant: The study highlights a substantial and consistent difference in how often decompressive craniectomy (DC) is used across different neurosurgical centers, even for patients with similar prognoses. This variation suggests a lack of clear consensus or strong evidence guiding the choice between DC and craniotomy.
- No Functional Outcome Difference: Despite the wide variation in practice, the study found no significant difference in functional outcomes at 6 months between patients treated with primary DC versus craniotomy when analyzed using instrumental variable methods. This is a crucial finding for clinical practice.
- DC Associated with More Complications: While functional outcomes were similar, patients undergoing primary DC were more likely to have follow-on surgeries and complications.
- Instrumental Variable Analysis is Key: The use of IV analysis was critical in addressing the inherent confounding in observational studies, allowing researchers to draw more robust conclusions about treatment effectiveness by leveraging natural variations in practice.
- Evidence Supports Restricting Primary DC: The findings suggest that primary DC should be reserved for specific situations, such as when severe intraoperative brain swelling prevents the immediate replacement of the bone flap, rather than being used as a routine preemptive measure.
Key Results
- Out of 4559 patients enrolled in CENTER-TBI, 336 underwent acute surgery for ASDH. Of these, 91 (27%) received primary DC and 245 (73%) received craniotomy.
- The proportion of primary DC within total acute surgeries ranged from 6% to 67% across 46 centers, with an interquartile range of 12-26%.
- The odds of receiving a DC for prognostically similar patients in one center versus another were trebled (adjusted median odds ratio 2.7, p < 0.0001).
- Primary Outcome: Higher center preference for DC was not associated with better functional outcome. For every 14% increase in DC use per center, the adjusted common odds ratio (OR) for better functional outcome was 0.9 (95% CI 0.7-1.1), indicating no benefit and a slight suggestion of harm.
- Secondary Outcomes: Primary DC was associated with more follow-on surgeries and complications (e.g., secondary cranial surgery: 27% vs. 18%; shunts: 11% vs. 5%). In-hospital mortality was similar between the groups.
- The study found that primary DC was associated with a worse clinical course, even though functional outcomes were similar.
Limitations and Future Work
- Sample Size for Certain Analyses: While the overall cohort was large, specific subgroup analyses (e.g., for quality of life measures) were limited by small numbers.
- Generalizability: The study population was primarily from tertiary referral centers, which may limit generalizability to other hospital settings.
- Residual Confounding: Despite the use of IV analysis and adjustments, the authors acknowledge the possibility of residual confounding from other local practice variations.
- Interpretation of Individual Effects: The IV analysis provides insights into the effect of changing center policy, but it's difficult to extrapolate these findings to predict the outcome for an individual patient.
- Future Work: The authors suggest that guidelines should restrict primary DC to patients with severe brain swelling to promote consistency in practice. They also emphasize the value of collecting granular data to better understand clinical courses and the potential of observational comparative effectiveness research. Further research could explore the specific factors driving the observed practice variation and refine methods for addressing confounding in neurotrauma research.
Summary updated