[Bedside ultrasound monitoring of optic nerve sheath diameter is a predictive factor for 28-day coma, delirium and death in etiologically diverse critically ill patients].

Objective: To explore whether the optic nerve sheath diameter (ONSD) within 24 hours of intensive care unit (ICU) admission is the predictor of 28-day delirium or coma and death in etiologically diverse critically ill patients.

Methods: A prospective, observational study was conducted. The critically ill patients admitted to the emergency ICU of Cangzhou Central Hospital from January 2021 to October 2022 were enrolled. Bedside ultrasound monitoring ONSD was performed within 24 hours of ICU admission. The consciousness status was assessed daily during ICU hospitalization. Coma was defined as Glasgow coma scale (GCS) score < 8 or Richmond agitation-sedation scale (RASS) score -4 or -5. Delirium was defined as responsiveness to verbal stimulation and with a positive confusion assessment method-intensive care unit (CAM-ICU). A positive result of CAM-ICU was defined as acute change or fluctuating course of mental status+inattention+altered level of consciousness or disorganized thinking. X-tile software analysis was used to visualize the best cut-off value for creating divisions in predicting 28-day coma or delirium and death, and then Kaplan-Meier curves were plotted. ONSD≥the optimal cut-off value from X-tile analysis was defined as ONSD broadening. ONSD broadening and related indicators were enrolled, and multivariate Cox regression analysis was used to analyze the risk factors of 28-day coma or delirium and 28-day death in etiologically diverse critically ill patients.

Results: A total of 321 critically ill patients were enrolled. Of them, 49 had primary brain injury, 54 had hypoxic ischemic brain injury (HIBI) after cardiac arrest, 70 had acute heart failure, 73 had sepsis, and 75 had other causes. Coma affected 184 patients (57.3%), and delirium affected 173 patients (53.9%). At 28 days of follow-up, 100 patients died, 16 patients remained comatose and 20 patients remained delirious. In all patients, as the GCS score decreased upon admission to the ICU, there was a gradually increasing trend in ONSD [GCS score 15 group: 5.20 (4.93, 5.43) mm, GCS score 10-14 group: 5.30 (4.90, 5.65) mm, GCS score 6-9 group: 5.40 (5.10, 5.80) mm, GCS score < 6 group: 5.70 (5.20, 5.96) mm, P < 0.05]. X-tile software analysis showed that in all patients and five etiological subgroups, ONSD broadening was a predictor for 28-day coma or delirium, and the optimal cut-off value was obtained (5.60 mm for all patients, 4.90 mm for primary brain injury, 5.75 mm for HIBI after cardiac arrest, 5.40 mm for acute heart failure, 5.90 mm for sepsis, and 5.75 mm for other causes). The Kaplan-Meier curves were plotted according to the optimal cut-off values, and the results showed that the higher the ONSD, the higher the incidence and duration of coma or delirium within 28 days in above patient population. X-tile software analysis showed that in all patients, and HIBI after cardiac arrest, sepsis and other causes patients, ONSD was a predictor for 28-day death, and the optimal cut-off value was obtained (6.20 mm for all patients, 5.85 mm for HIBI after cardiac arrest, 5.35 mm for sepsis, and 6.10 mm for other causes). The Kaplan-Meier curves were plotted according to the optimal cut-off values, and the results showed that the higher the ONSD, the higher the 28-day survival rate and the shorter survival duration in above patient population. Multivariate Cox regression analysis showed that ONSD broadening was an independent risk factor for 28-day coma or delirium in all patients [hazard ratio (HR) = 1.513, 95% confidence interval (95%CI) was 1.093-2.095, P = 0.013] and patients with primary brain injury (HR = 5.739, 95%CI was 2.112-15.590, P = 0.001). However, ONSD broadening was not independently associated with 28-day death in all patients or in the five etiological subgroups.

Conclusions: ONSD within 24 hours of ICU admission is an independent risk factor for 28-day coma or delirium in etiologically diverse critically ill patients. It serves as a predictor for 28-day coma or delirium in 5 subgroups of etiology including primary brain injury, HIBI after cardiac arrest, acute heart failure, sepsis, and other causes, but not for 28-day death.