Respiratory Care

Expert management of tracheal intubation has become fundamental to the routine practice of pulmonary physicians who work in respiratory intensive care units (ICUs). In Italy, tracheal intubation is not included as part of the training in respiratory medicine, and pulmonary physicians are usually dissuaded from managing intubations.

We prospectively studied the intubation success rate in 46 consecutive respiratory ICU patients who required either emergency or urgent intubation, conducted by 3 intubation-trained pulmonary physicians in our respiratory ICU. Intubation success was defined as successful tracheal intubation without any of 7 pre-defined complications.

There were 17 emergency intubations and 29 urgent intubations. Intubation was successful in 43 of the 46 intubation attempts. Complications occurred in 3 cases: 2 patients needed to be intubated by an anesthesiologist, and 1 patient received fiberoptic intubation.

Pulmonary physicians trained in tracheal intubation can have a high success rate in performing intubation in the respiratory ICU. Collaborative efforts between anesthesiologists and pulmonary physicians are necessary to optimize the training, skill-retention, and back-up for advanced airway management in the respiratory ICU.

The time course of the physiological derangements that result from ventilatorinduced lung injury has not been adequately described. Similarly, the regional topographies of pleural pressure and tissue edema have not been carefully mapped for this injury process.

Lung injury was induced in 9 normal pigs by ventilating for 6 hours at a transpulmonary pressure of 35 cm H₂O, with the animals in the supine position. Eight additional normal pigs received right thoracotomy to place pleural-surface-pressure sensors prior to an identical period and intensity of injurious ventilation. Gas exchange and lung mechanics were tracked in all the animals. Cytokines (tumor necrosis factor alpha, interleukin 6, and interleukin 8) in peripheral blood were assayed at 2 hour intervals, beginning at the onset of mechanical ventilation, from all the animals.

After a brief “induction” period, P_aO2 and tidal volume declined steadily in the animals that were ventilated to induce lung injury. The rate of decline was greater in the animals that received thoracotomy. The pleural pressure gradient steadily increased from ventral to dorsal. The serum cytokine levels did not evolve with developing injury, but cytokines were elevated at the onset of ventilation. Tissue edema, as assessed by the ratio of wet weight to dry weight, was greater in the thoracotomized animals than in the nonthoracotomized animals, and tissue edema tended to be greater in the caudal lung regions than in the cephalad lung regions.

Following the induction period, the development of ventilator-induced lung injury progressed steadily and then plateaued, as assessed by quantitative physiology variables during 6 hours of ventilation at a transpulmonary pressure of 35 cm H₂O. Greater injury developed in animals that had a coexisting potential insult (thoracotomy). Injury development was not paralleled by bloodborne inflammatory cytokines.

Noninvasive positive-pressure ventilation (NPPV) delivers air at a high flow, which is associated with airway mucosal drying and impaired airway functioning.

To examine the effects of mechanical ventilation parameters on relative humidity and absolute humidity during NPPV, and to evaluate the effect of a heated passover humidifier on relative humidity, absolute humidity, and ventilator performance during NPPV.

We performed a bench study to assess the effects of inspiratory positive airway pressure (IPAP) of 10 cm H₂O, 15 cm H₂O, and 20 cm H₂O, respiratory rates of 12 breaths/min and 24 breaths/min, and inspiratory-expiratory ratios of 1:2 and 1:3 on relative and absolute humidity. The measurements were obtained on room air and with a heated humidifier at medium and maximum heater settings.

Without humidification, the relative humidity in the NPPV circuit (range 16.3–26.5%) was substantially lower than the ambient relative humidity (27.6–31.5%) at all ventilatory settings. Increasing the IPAP decreased the relative humidity (Spearman's rho = 0.67, p < 0.001). Changing the respiratory rate or inspiratory-expiratory ratio had no significant effect. Both relative and absolute humidity increased with humidification, and the air was fully saturated at the maximum heater setting. Delivered IPAP was reduced by 0.5–1 cm H₂O during humidification.

NPPV delivers air with a low relative humidity, especially with high inspiratory pressure. Addition of a heated humidifier increases the relative and absolute humidity to levels acceptable for nonintubated patients, with minimal effect on delivered pressure. Consideration should be given to heated humidification during NPPV, especially when airway drying and secretion retention are of concern.

With a high-frequency percussive ventilator and a mechanical lung model, to measure tidal volume (V_T), pulsatile pressure amplitude (difference between peak and nadir pulsatile pressure [ΔP]), and mean airway pressure (P̄_aw) at various pulsatile frequencies, pulsatile inspiratory-expiratory ratios (I:E_p), and pressures (measured at the interface between the pulse-generator and the endotracheal tube [P_vent]).

With the endotracheal tube inside an artificial trachea, we manipulated the high-frequency percussive ventilation settings and adjuncts, including pulsatile frequency, I:E_p, and P_vent by manipulating pulsatile flow. We also studied the effects of partially deflating the endotracheal tube cuff. We measured P̄_aw, pulsatile pressure amplitude at the carina (ΔP_c), and pulsatile V_T at the carina. With the cuff partly deflated, we measured the fraction of inspired oxygen (F_IO2 ) in the gas efflux above and below the cuff.

Increasing the pulsatile frequency from 300 cycles/min to 600 cycles/min and changing the I:E_p from 1:3 to 1:1 significantly reduced V_T (p < 0.001). P̄_aw and ΔP_c were unaffected by the change in pulsatile frequency or I:E_p, except when we did not preserve the pulsatile flow. The measured V_T range was from 19.1 mL (at 600 cycles/min) to 47.3 mL (at 300 cycles/min). Partial cuff deflation did not significantly reduce P̄_aw or ΔP_c, but it did significantly reduce V_T and F_IO2 .

During high-frequency percussive ventilation, the pulsatile frequency is inversely related to V_T. Partial cuff deflation reduces the delivered F_IO2 .

High-frequency ventilation can be delivered with either oscillatory ventilation (HFOV) or jet ventilation (HFJV). Traditional clinician biases may limit the range of function of these important ventilation modes. We hypothesized that (1) the jet ventilator can be an accurate monitor of mean airway pressure (P̄_aw) during HFOV, and (2) a mathematical relationship can be used to determine the positive end-expiratory pressure (PEEP) setting required for HFJV to reproduce the P̄_aw of HFOV.

In phase 1 of our experiment, we used a differential pressure pneumotachometer and a jet adapter in-line between an oscillator circuit and a pediatric lung model to measure P̄_aw, PEEP, and peak inspiratory pressure (PIP). Thirty-six HFOV setting combinations were studied, in random order. We analyzed the correlation between the pneumotachometer and HFJV measurements. In phase 2 we used the jet as the monitoring device during each of the same 36 combinations of HFOV settings, and recorded P̄_aw, PIP, and ΔP. Then, for each combination of settings, the jet ventilator was placed in-line with a conventional ventilator and was set at the same rate and PIP as was monitored during HFOV. To determine the appropriate PEEP setting, we calculated the P̄_aw contributed by the PIP, respiratory rate, and inspiratory time set for HFJV, and subtracted this from the goal P̄_aw. This value was the PEEP predicted for HFJV to match the HFOV P̄_aw.

The correlation coefficient between the pneumotachometer and HFJV measurements was r = 0.99 (mean difference 0.62 ± 0.30 cm H₂O, p < 0.001). The predicted and actual PEEP required were highly correlated (r = 0.99, p < 0.001). The mean difference in these values is not statistically significantly different from zero (mean difference 0.25 ± 1.02 cm H₂O, p > 0.15).

HFJV is an accurate monitor during HFOV. These measurements can be used to calculate the predicted PEEP necessary to match P̄_aw on the 2 ventilators. Replicating the P̄_aw with adequate PEEP on HFJV may help simplify transitioning between ventilators when clinically indicated.