Intubation and DNI/DNAR practices in elderly patients dying early from severe infection

Abstract Background The guidelines for treatment of patients with severe infection or sepsis do not include the indications for endotracheal intubation, nor do they address DNAR/DNI orders. Methods We aimed to describe the rates, characteristics and association (or lack thereof) between tracheal intubation and DNI/DNAR orders among elderly patients with severe infection or sepsis who died within 48 h of hospital admission. We conducted a single-center, retrospective, case-controlled study at a tertiary medical center. Adult patients aged ≥ 65 years with severe infection or sepsis who died within 48 h of admission were included. Data on demographics, comorbidities, DNAR/DNI status, intubation, and end-of-life (EoL) discussions were collected and analyzed. Results Of the 304 patients included, 29.9% underwent tracheal intubation. Only 14.5% of patients (44/304) had pre-existing DNAR/DNI orders, while 71.1% received such orders within 48 h. The presence of DNAR/DNI orders was strongly associated with lower intubation rates (p < 0.0001). However, 12.2% of patients with a DNAR/DNI order from the current hospitalization were intubated. Gaps in documentation were common: 18.1% had no recorded EoL discussions, and in many cases, EoL discussions occurred only after intubation. Conclusions In elderly patients with severe infection or sepsis who died within 48 h of admission, tracheal intubation rates were higher than previously reported, while pre-existing DNAR/DNI orders were rare despite a high comorbidity load. DNAR/DNI orders were associated with reduced intubation rates but were inconsistently followed, especially pre-hospital. Poor timing and documentation of end-of-life discussions suggest a need for earlier, structured communication and better integration of patient wishes across care settings.