LAUSR

Optimal care when the end of life is near is based on in the principle of patient autonomy. Surgeons, however, are often hesitant to limit life-sustaining interventions even if patients prefer less aggressive care at the end of life.1,2 The challenges for incorporating patient preferences and values in the context of high-risk surgery may be explained in several ways. Schwarze et al2 described the concept of surgical “buy-in,” in which surgeons performing high-risk surgery expect their patients to agree to certain postoperative care in an unspoken contractual agreement made during the preoperative conversation. The expectations of buy-in are grounded in surgeons’ strong sense of responsibility for surgical outcomes and the pervasive notion that a patient who dies represents a treatment failure. Second, the pressure of the 30-day mortality metric may create an incentive to encourage aggressive lifesustaining treatments.3 In the study published in this issue of JAMA Surgery, Lilley et al4 examine end-of-life care among elderly Medicare beneficiaries treated for traumatic brain injury at lowand high-mortality hospitals. They found that end-of-life care for patients who died in the hospital was similar across institutions. Patients who died after being discharged from lowmortality hospitals, however, were less likely to have a tracheostomy and gastrostomy and more likely to enroll in hospice. Furthermore, patients discharged alive from lowmortality hospitals were less likely to die within 30 days of discharge. This study has limitations, including several mentioned by the authors, such as the retrospective design and limited variables available for analysis. Perhaps one of the most important limitations in assessing benefits of hospice enrollment is that no information was available about prior advanced care planning or input from surrogate decision makers. It therefore remains unclear if patients were able to have end-of-life care that fit with their preferences. Surgical care in seriously ill patients is challenging and necessitates better communication between patients and their surrogates, surgeons, and palliative care professionals. Invasive life-sustaining treatments at the end of life are common: a national study of decedent Medicare & Medicaid beneficiaries found that 18.3% underwent a procedure within the last month of life.5 Several studies, including this study,4 have demonstrated that hospice does not compromise survival, underscoring the fact that the fear that hospice enrollment may compromise surgical outcomes is not warranted.6 Surgeons have a unique perspective on the potential benefits and harms of intensive treatment for patients with advanced illnesses and are thus well positioned to help patients and their surrogates determine how invasive interventions do or do not fit with their patients’ goals for end-of-life care.