Gantner D, Wiegers E, Bragge P, Finfer S, Delaney A, van Essen T, Peul W, Maas AIR, Cooper DJ. Decompressive Craniectomy Practice following Traumatic Brain Injury in Comparison with Randomized Trials: Harmonized, Multi-Center Cohort Studies in Europe, the United Kingdom, and Australia. J Neurotrauma. 2022 Apr 6. doi: 10.1089/neu.2021.0312. Epub ahead of print. PMID: 35243877.
Background: Severe traumatic brain injury (TBI) is highly disabling, with over 50% of patients facing mortality or long-term disability. Secondary decompressive craniectomy (DC) is a treatment supported by Brain Trauma Foundation Guidelines based on data from the DECRA and RESCUEicp clinical trials. These studies showed that while DC reduces ICP, it may increase the number of survivors with severe disabilities without improving overall outcomes. The authors of this study hypothesized that contemporary international use of DC may vary between regions and may not be aligned with evidence.
Methods: In this cohort study, data on patients with severe TBI that were prospectively collected between 2015 and 2017 were pooled from 54 hospitals in 19 countries, including data from across the EU, UK, and Israel (CENTER-TBI), and data from two centers in Australia (OzENTER-TBI). Clinical variables that were considered included patient demographics, injury mechanisms and severity scores (Marshall classification), the presence of pre-hospital hypoxia and hypotension, initial GCS scores, emergency and ICU therapies, details related to DC, and non-operative therapies for ICP including sedation, osmotherapy, muscle relaxants, therapeutic hypothermia, hypocapnia, and CSF drainage. Poor functional outcome was defined as an Extended Glasgow Outcome Score (GOSE) <5, and the frequency of patients’ alignment with eligibility criteria for the DECRA and RESCUEicp trials was also reported. DECRA patients were those with GCS 3-8, diffuse injury (Marshall CT class I-IV), and persistent ICP >20 mmHg that was refractory to non-operative therapies within 72 hours of injury. RESCUEicp patients included those with diffuse injury or traumatic mass lesions with persistent ICP >25 mmHg that was refractory to non-operative therapies at any time during a patient’s ICU stay. The primary outcome of interest was the proportion of patients who underwent secondary DC for refractory ICP, with analyses stratified by TBI occurring before or after January 1, 2017 (as the RESCUEicp trial was published in September 2016).
Patient population: In total, 2336 patients were included, of whom 80% were from the EU, 11.6% from the UK, and 8.5% from Australia, with 45.7% of all patients suffering a severe TBI. Median age was 49 years and 73% were men. Median GCS on admission was lower in the UK (6) compared to Australia (8) and the EU (10). Sixty-nine percent of patients overall had diffuse injury, and metabolic suppression was used as part of ICP treatment in 25% of patients. The rates of traumatic intracranial lesions were higher in the UK (39%) compared to the EU (30%) and Australia (30%). Fewer patients received ICP monitors in the EU (41%) than in the UK (64%) or Australia (59%). Poor outcomes at 6 months were observed in 44.5% of all patients.
Use of DC and alignment with evidence: The proportions of patients who met enrollment criteria for DECRA and RESCUEicp were both low (5% and 10%, respectively). Patients with traumatic mass lesions underwent DC more frequently than those with diffuse injury (30.5% vs. 4.5%). The frequency of DC was highest in the UK compared to the EU and Australia, for both patients with mass lesions (49% vs. 26.5% vs. 40%) and diffuse TBI (7% vs. 5% vs. 1%), which aligned with the greater severity of injury in the UK cohort. The rates of secondary DC were highest in the UK compared to the EU and Australia (22% vs. 8% vs. 6%). A unilateral fronto-temporal approach was the most common surgical approach (75%) in both patients with mass lesions (80%) and diffuse injury (69%). In adjusted analyses, rates of secondary DC were lower in older patients (OR 0.97 per year; 95% CI 0.96-0.98; p<0.001), those with less severe TBI based on initial GCS (OR 0.9 per GCS point; 95% CI 0.85-0.95; p<0.001), and those with diffuse injury (OR 0.31; 95% CI 0.19-0.51; p<0.001), while rates were higher in patients treated in the UK (OR 2.14; 95% CI 1.21-3.78; p<0.009). Only a small percentage of patients who underwent secondary DC would have met DECRA or RESCUEicp eligibility criteria (9% in the UK, 9.5% in the EU, and 0% Australia for DECRA; 13.6% in the UK and EU, 0% in Australia for RESCUEicp). Furthermore, the publication of RESCUEicp in September 2016 did not change the selection criteria utilized as there was no difference in the proportion of eligible patients undergoing DC before and after publication (26.7% before vs. 27% after, p=0.506)
Commentary: In this large harmonized multi-center cohort study evaluating real world experiences in the utilization of DC, there was a high degree of variability in the proportions of patients who underwent secondary DC between the EU, UK and Australia. Most patients who underwent DC would not have met enrollment criteria for the DECRA and RESCUEicp trials, suggesting that DC decisions are yet to be informed by high quality evidence. Decompressive craniectomy post DECRA and RESCUEicp is mired in debate and controversy. While often life-saving, DC is associated with a higher likelihood of long-term unfavorable outcomes in survivors, and decision-making must therefore include discussions with patients’ surrogates about potential outcomes, recovery times, and quality of life.
In this international cohort, few patients met eligibility criteria for DECRA or RESCUEicp (5% and 19%, respectively). Many patients would have been ineligible for these trials due to having received a primary DC (47%), not developing ICP crises that were refractory to medical therapies (39%), or being outside age criteria (19%). Patients with diffuse TBI were less likely to receive DC than mass lesions. A large proportion of patients in this cohort underwent primary DC prior to non-operative therapies. Patients in the UK were younger and had more severe injuries, perhaps explaining their higher likelihood for secondary DC. Comparative effectiveness studies that enroll older patients and also examine primary DC are needed to further inform clinical practice, particularly in circumstances where DC is currently being used.
Prisma Health- Upstate
University of South Carolina School of Medicine-Greenville