Withdrawal of care: a 10-year perspective at a Level I trauma center.

BACKGROUND

Withdrawal or limitation of care (WLC) in trauma patients has not been well studied. We reviewed 10 years of deaths at our adult Level I trauma center to identify the patients undergoing WLC and to describe the process of trauma surgeon-managed WLC.

METHODS

This is a retrospective review of WLC. Each patient was assigned to one of three modes of WLC: care withdrawn, limited or no resuscitation, or organ harvest. Frequency, timing, and circumstances of WLC, including family involvement, ethics committee consultation, palliative care, and hospice, were reviewed.

RESULTS

From 2000 through 2009, 375 patients died with WLC (54% of all deaths; 93% at ≥ 24 hours). For age ≥ 65 years, 80% were WLC. Overall, 15% had advance directive documents. Traumatic brain or high cervical spine injury was the cause of death in 63%. Factors associated with WLC included age, comorbidities, injury mechanism and severity, and nontrauma activation status. At time of death, 316 (84%) WLC were under trauma surgeon management. In this group, mode of WLC was care withdrawn in 74%, organ harvest in 20%, and limited or no resuscitation in 6%. Rationale for WLC in non-organ harvest patients was poor neurologic prognosis in 86% and futility in 76%. When family was identified, end-of-life discussions with physicians occurred in 100%. Conflicts over WLC occurred in 6.6% and were not associated with any demographic group. Ethics committee was involved in 2.8%. For care-withdrawn patients, median time to death from first WLC order was 6.6 hours. Palliative care and hospice consults (6% and 9%) increased yearly.

CONCLUSIONS

WLC occurred in over 50% of all trauma deaths and exceeded 90% at ≥ 24 hours. Hospice and palliative care were increasingly important adjuncts to WLC. Guidelines for WLC should be developed to ensure quality end-of-life care for trauma patients in whom further care is futile.

LEVEL OF EVIDENCE

III, therapeutic study.