Introduction: Liberation from long-term mechanical ventilation (MV) during tracheostomy collar trials (TCT) is often confounded by both physiological and psychological factors. Discerning dyspnea from anxiety remains challenging due to symptom overlap. As a result, patients may either be returned to MV prematurely or pushed beyond physiological limits based on subjective assessments, arbitrary time goals, or clinical bias. Lung electrical impedance tomography (EIT) provides real-time, regional lung ventilation data and may offer objective insight into TCT failure, although it remains underutilized in this context. Description: A 54-year-old female with a history of severe COPD (home O2), hypertension, hyperlipidemia, and generalized anxiety disorder was admitted following a motor vehicle collision with polytrauma involving the head, chest, and lower extremities. After prolonged MV, a tracheostomy was performed (day 12); TCT attempts began on day 16. Over subsequent days, multiple TCTs showed minimal progress, with discontinuation guided by time-based endpoints. During the 5th TCT, EIT was employed, revealing marked unilateral ventilation asymmetry and significant left-to-right intrapulmonary gas shift (pendelluft), indicative of severe regional inhomogeneity. EIT also showed a progressive decline in end-expiratory lung impedance (EELI) and increasing ventilation heterogeneity over one hour. At minute 45, the patient developed visible paradoxical breathing, tachypnea, accessory muscle use, and reported intense dyspnea and anxiety. She was subsequently returned to MV. Discussion: Previous TCTs were guided by fixed time ratios (e.g., 4:2 on:off MV), without individualized physiological monitoring. Documentation noted anxiety and respiratory distress, but anxiety was often treated as a psychosomatic overlay rather than a primary sign of respiratory compromise. In this instance, EIT provided objective evidence supporting discontinuation of the TCT based on physiological deterioration, rather than subjective distress alone.
Ring et al. (Sun,) studied this question.