The Potential Risks of Pressure Support Ventilation

Abstract

Pressure support ventilation (PSV) is the most commonly used mode of assisted ventilation in the intensive care unit. Setting the ventilator in PSV mode appears to be easy, yet optimizing patient-ventilator interaction in terms of matching a patient’s ventilatory demands with the level of assist, and the neural inspiratory time (Ti) with the mechanical Ti, is particularly challenging. Excessive assist can often occur during PSV, leading to weak or ineffective efforts and periodic breathing, promoting diaphragmatic injury and prolonged mechanical ventilation. Additionally, mismatch between mechanical and neural Ti almost invariably occurs, leading to expiratory asynchronies and prohibiting accurate measurement of driving pressure, placing patients with high respiratory drive at risk of lung injury. Three types of problem hindering patient-ventilator interaction in PSV can be identified. First, factors inherent to the operation of PSV, which include the limited dependence, and thus matching, between mechanical Ti and neural Ti, and the delivery, after ventilator triggering and independently of the patient’s effort, of a minimum tidal volume, depending on the level of pressure support. Second, the response pattern of the control-of-breathing system to changes in ventilatory demands, which is mediated mainly by changes in effort and minimally by changes in respiratory rate. Lastly, the lack of clinical signs of excessive ventilatory assist, and automated monitoring tools, make optimization of patient-ventilator interaction during PSV difficult. Understanding the challenges of PSV can facilitate prompt recognition and management of the problems it may create, and help provide lung- and diaphragm-protective ventilation during PSV.