Supplementary file 1_Nurse-performed diaphragm ultrasound integrated with spontaneous breathing trial criteria for risk stratification of extubation outcomes in neurosurgical critically ill patients: a multicenter prospective cohort study.docx

BackgroundIn neurosurgical intensive care units (ICUs), conventional spontaneous breathing trial (SBT) criteria may not adequately reflect diaphragm function, which can contribute to premature or delayed extubation. Diaphragm ultrasound provides objective bedside measures to support individualized weaning.
MethodsThis multicenter prospective cohort study enrolled 188 patients from six tertiary hospitals. Patients were sequentially managed in three phases: control (conventional SBT), diaphragm thickening fraction (DTF)–guided management (extended SBT if DTF was below the prespecified threshold), and combined DTF plus diaphragmatic excursion (DE)–guided management. Primary outcomes included first SBT success, first extubation success, reintubation, and duration of mechanical ventilation. Receiver operating characteristic (ROC) and Cohen’s kappa analyses were used to evaluate predictive accuracy and agreement between SBT results and extubation outcomes.
ResultsNo significant differences were observed in first SBT success or first extubation success (p = 0.127; p = 0.900). Reintubation occurred significantly more often in the control group than in the DTF and DTF + DE groups (p = 0.004). Total ventilation time was also longer in the control group (p < 0.001). ROC analysis demonstrated stronger predictive value for DTF than for DE, with limited incremental benefit from combining the two measures (p < 0.01). Kappa analysis showed improved agreement between SBT results and extubation outcomes when diaphragm-based stratification was applied (p < 0.01).
ConclusionNurse-performed diaphragm ultrasound can be implemented to support physician-led weaning assessment. DTF-based stratification was associated with reduced reintubation and shorter ventilation duration.