There is no evidence that spontaneous breathing trials train the patient and his or her muscles. In fact, it is unlikely that spontaneous breathing trials result in increasing the strength of respiratory muscles. It has been clearly shown that, if intensity of load is not monitored and guaranteed, no training effect can be expected [6]. Spontaneous breathing trials do not improve the patient’s condition but provided information about how to continue. Spontaneous breathing capacity reflects the impact of the disease and the necessity of mechanical ventilation. Frequently, the spontaneous breathing capacity displayed is much better than anticipated before the trial is performed. This allows more rapid weaning to proceed. Impacts on pCO₂ and dyspnea reveal whether spontaneous breathing is safe or risks resulting in muscular fatigue if unduly prolonged. The evolution of oxygen saturation during spontaneous breathing reveals whether hypoxemia can be managed without intubation. In addition, physical and behavioral ability to cooperate and tolerate an unpleasant respiratory load without panic or anxiety can be tested during a spontaneous breathing trial. Some patients who have experienced life-threatening episodes during their intensive care stay may become anxious about any change that is made. Explaining the process to the patient and staying him or her, asking about discomfort, and terminating the trial as soon as adverse events occur may be prerequisites for success.

There are two approaches to spontaneous breathing trials: (a) breathing spontaneously while on the ventilator, with minimum support by continuous positive airway pressure (CPAP) and pressure support, and (b) T-tube breathing, with the tube in place but not connected to the ventilator and with an oxygen adaptor inserted to apply additional oxygen as required.

Supporters of the CPAP trial argue that the patient is still monitored by the ventilator. However, as previously mentioned, the spontaneous breathing trial should be monitored by an experienced therapist to recognize and avoid any adverse event and to motivate the patient. Thirty minutes of spontaneous breathing have been shown to be sufficient, with no additional information gained if the trial is prolonged beyond that point [7]. A well-supervised T-piece trial does not require any monitoring by the ventilator. Perfect interaction with the ventilator during CPAP and pressure support can be difficult to accomplish in a spontaneous breathing, swallowing, and coughing subject, and we worry that breathing against the machine may impair the patient. Additionally, ventilator support, although minimal, may mask some consequences of spontaneous breathing [9, 10] because of muscle overload or noncompensable hypoxemia and, as a consequence, the risk for reintubation. In severe COPD, CPAP must be kept low, at least below the intrinsic PEEP, if increased hyperinflation might further aggravate. Because intrinsic PEEP is variable with time and not homogenously distributed in the lung, we prefer the T-tube trial [11].

As soon as the patient is extubated and the endotracheal tube is replaced by a nasal or naso-oral interface, mechanical ventilation can be performed intermittently according to the patient’s spontaneous breathing capacity. Until complete stabilization is reached, spontaneous breathing periods may be kept short to avoid exhaustion and rapid shallow breathing followed by difficulties to readapt to noninvasive ventilation. Thereafter, prolongation of spontaneous breathing and reduction of the time on the ventilator may follow the overall improvement. Noninvasive ventilation, in contrast to endotracheal ventilation, allows the patient to be transferred to the normal ward, to a rehabilitation center, or to home even if the patient is not completely weaned. Sufficient care and cooperation to continue with intermittent ventilation and stability are the only requirements (Table 41.2).

Noninvasive ventilation can be successfully performed in the totally ventilator dependent patient. However, the more spontaneous breathing is preserved, the lower the risk for reintubation. Thirty minutes of a stable T-piece trial with stable pCO₂ and oxygen saturation should be achieved prior to extubation and transfer to noninvasive ventilation. However, with sufficient expertise and the support of well-trained specialists, most of our patients are extubated after a shorter T-piece trial [5]. Full cooperation and perfect adaptation are the major determinants of failure or success [12]. If the patient is fully cooperative and adapts well to invasive tube ventilation without the need of sedatives, mask ventilation can be expected to work equally effectively. Uncooperativeness or the necessity to apply sedatives considerably increase the risk of failure.