Treatment-Limiting Decisions Occur Frequently, Early in Patients With Severe TBI

Treatment-limiting decisions, half of which were made within 3 days of admission, occurred in nearly half of patients with severe traumatic brain injury (TBI) and occurred more frequently over time, according to study results published in Brain and Spine.

Researchers conducted a retrospective cohort study between 2008 and 2017 at 2 level 1 trauma centers to examine the prevalence, timing, and considerations for treatment-limiting decisions in patients with severe TBI. Patients with severe TBI and Glasgow Coma Scores (GCS) between 3 and 8 on admission were eligible for inclusion. Baseline probability of mortality and unfavorable outcome at 6 months was calculated using the Corticosteroid Randomisation after Significant Head Injury (CRASH) prognostic model. Treatment-limiting decisions were classified as withholding life-saving interventions or withdrawing life-sustaining measures and further distinguished as withholding intensive care unit (ICU) admission, neurosurgical intervention, escalation of intracranial pressure targeted treatment, or organ-support; do not resuscitate order; or withdrawing organ-support, intracranial pressure targeted treatment, or nutrition support. An analysis of variance test was used to test differences in the CRASH baseline prognosis score over years.

A total of 270 patients were included in the final analysis. Treatment-limiting decisions were made in 117 (43%; mean age, 58; men, 67.5%; median GCS, 3) patients, 95.9% of whom died. The CRASH prognosis risk score of 14-day mortality and unfavorable outcome at 6 months was 72.6% and 70.6% for the groups in which treatment-limiting decisions were and were not made, respectively (P =.71).

The percentage of treatment-limiting decisions in patients with severe TBI increased from 20.0% in 2008 to 42.9% in 2012 and 64.3% in 2017; however, the CRASH prognosis risk score of unfavorable outcome at 6 months did not significantly differ over time (2008: 73.1%; 2012: 73.9%; 2017: 72.5%).

Treatment was initiated, but later withdrawn in 68.4% of patients. In the other 31.6% of patients, treatment was withheld at presentation. Withdrawal of organ-support (64.1%) and withholding neurosurgical intervention (29.9%) were the 2 most frequent types of treatment-limiting decisions.

Treatment-limiting decisions were primarily made at the ICU (78.6%), followed by the ward (12.8%), and emergency room (8.5%). The median GCS at time of treatment-limiting decision was 3 and pupillary abnormalities were observed in 82.9% of patients.

Median time from admission to treatment-limiting decision was 2 days. The majority (50.4%) of treatment-limiting decisions were made within 3 days of admission, 28.2% of which were made within 1 day of hospitalization. Median time from treatment-limiting decision to death was 1 day.

The primary reason for a treatment-limiting decision (66.7%) was that the patient was deemed unsalvageable, followed unacceptable quality of life (QOL) as presumed by the physician (16.2%) or proxy (14.5%). Nearly all (99.1%) treatment-limiting decisions were made by a multidisciplinary team with a large percentage involving the proxy (75.2%).

The main reason for death in the group of patients in which a treatment-limiting decision was and was not made was the initial intracranial injury (91.5% vs 57.1%, respectively), followed by lesser reasons such as extracranial injury (1.8% vs 14.3%) and medical complications (6.0% vs 14.3%).

Study limitations include the retrospective design, use of potentially outdated data from 2008 to 2017, reduced generalizability of results to other geographic regions, and lack of data on radiographic characteristics at timing of treatment-limiting decisions.

“Future research should address whether prognostic nihilism contributes to self-fulfilling prophecies,” the study authors concluded.