Long-term frailty trajectory in older ICU survivors requiring mechanical ventilation: a prospective cohort study in Norway.

BACKGROUND: While intensive care unit (ICU) capacity remains constrained, the number of older patients receiving resource-demanding ICU treatment continues to rise. Yet, long-term outcomes and post-critical illness frailty trajectories remain poorly characterized. We examined long-term mortality and frailty trajectories, and their associations with clinical characteristics and ICU treatment intensity. METHODS: In this prospective, observational study across five Norwegian hospitals representing all health regions, we included patients aged ≥65 years who required invasive mechanical ventilation for ≥24 h. Frailty was assessed using the Clinical Frailty Scale (CFS) at baseline and at 3 and 12 months, categorized as robust (CFS 1-3), prefrail (CFS 4) and frail (CFS ≥ 5). Multiple regression models were used to examine associations between frailty trajectories and mortality, with baseline frailty, clinical characteristics and potentially modifiable aspects of ICU treatment. RESULTS: 346 patients were included. Mean age was 74 years (± 6), and 98 % lived at home before admission. The median ICU length of stay was 8 (IQR 9) days. Total mortality was 52 %, with most deaths occurring before hospital discharge. In the survivors, the prevalence of frailty increased from 23 % (79/346) at baseline, to 47 % (81/174) at three months. Between the three- and 12- month follow-up, CFS scores remained unchanged in 71 % (108/153), improved in 19 % (29/153) and deteriorated in 10 % (15/153). By 12 months 86 % (132/153) had returned home, though 32 % (49/153) required home care compared to 10 % pre-admission. Baseline CFS showed the strongest association with 12-month mortality (Risk Ratio 1.46, p < 0.001). Age (Risk Ratio 1.22, p < 0.001) and SOFA score on the first day after intubation (Risk Ratio 1.14, p < 0.01) were also significant predictors, whereas comorbidity was not. CONCLUSIONS: In elderly ICU-patients long-term frailty is deranged, and baseline CFS is closely associated with mortality. However, a small group of patients had a transitory decline in frailty at three months follow-up and then improved. Intervention studies targeting modifiable factors associated with long-term functional decline and mortality are warranted. TRIAL REGISTRATION: ClinicalTrials.gov (NCT06012942).