Poorer outcomes in patients with early postoperative delirium: 120-day follow-up of the Delirium Reduction by Analgesia Management in Hip Fracture (DRAM-HF) study.

Abstract

The Delirium Reduction by Analgesia Management-Hip Fracture (DRAM-HF) model of care, which incorporated a multicomponent intervention focussing on perioperative analgesia and medication optimisation, was associated with reduced Day 3 postoperative delirium (POD) amongst hip fracture patients. We investigated whether this effect was seen at 120 days postoperatively. We assessed 120-day outcomes in all patients who were included in the DRAM-HF study, by telephone, supplemented by electronic medical records, to include death (primary outcome), residential aged care facility (RACF) residence, patient/carer-reported frailty, hospital readmission and new dementia diagnosis. Amongst 300 patients (mean age 81.1, 70% female, none lost to follow-up), by 120 days, 8% (n = 24) had died; 25% of survivors (n = 68/276) were RACF residents. Twenty-two per cent were readmitted (n = 61/281). A new dementia diagnosis was reported by 6% (n = 17/281). Intervention status in the DRAM-HF trial (intervention/control) was not associated with death by 120 days (OR 0.83, 95% CI 0.36-1.93, p = 0.67) or other outcomes assessed. POD was independently associated with 120-day death (aOR 3.3, 95% CI 1.2-9.2, p = 0.02), RACF residence (aOR 2.2, 95% CI 1.1-4.7, p = 0.03) and patient/carer-reported frailty (aOR 5.6, 95% CI 1.0-30.7, p = 0.05), but not readmission (p = 0.21) or new diagnosis of dementia (p = 0.08). In this cohort, while the DRAM-HF bundle of care did not influence 120-day outcomes, patients who experienced POD had poorer clinical outcomes 120-day postfracture. Given that delirium was associated with death, RACF residence and frailty, models of care which have the potential to reduce POD may have benefits beyond the acute admission, and further investigation is needed.