Page 4
This usually implies relative hypovolemia and responds to intravenous fluid boluses. It is important to keep in mind that relatively “high” CVP or PCWP (e.g. mid to high 20’s) may indicate the patient is still hypovolemic because of the effect of the elevated airway pressures being transmitted to the measured intravascular pressures. This is especially true as mPaw exceeds values of 30 – 35 cmH20. If the patient is hypotensive, we will still generally administer serial fluid boluses (e.g. 500 ml normal saline or packed RBC’s if anemic) until the CVP or PCWP has clearly increased by 5 – 10 mmHg. At that point, if hypotension persists we add vasopressors (e.g. dopamine, epinephrine, norepinephrine) and reconsider the differential diagnosis of the hypotension. Pneumothorax It is important to realize that tension pneumothorax occurring during HFOV may not cause changes to the displayed mPaw or ∆P as the patient develops progressive hypotension and desaturation (10). A high index of suspicion is necessary for pneumothorax and confirmation (if time permits) requires an immediate portable chest radiograph. If in doubt, it is preferable to place a chest tube on the suspected side while waiting for the chest radiograph. It may be difficult by auscultation alone to detect the side of the pneumothorax - especially during HFOV because of the background noise of the ventilator and the diffuse transmission of airway sounds, however, the loss of chest wiggle that usually occurs on the affected side will provide man important physical clue. If a previously attempted central venous line (either internal jugular or subclavian vein) or thoracentesis is apparent, we usually will empirically place the chest tube on that side if time does not permit radiographic confirmation of pneumothorax. We discourage placing 14 gauge needles into the anterior thorax for suspected pneumothorax. If a pneumothorax was not present, the needle insertion has now likely caused one and will need immediate follow-up with a chest tube. It is always preferable to place a chest tube with direct palpation of the underlying lung to confirm placement of the tube in the pleural space. If a pneumothorax was not present then at least a bronchopleural fistula has not been created iatrogenically. To date there has been no convincing clinical evidence that the frequency of pneumothorax in adults is less or greater with HFOV compared with conventional ventilation. Several case reports and small series indicate that HFOV may provide successful oxygenation and ventilation in the presence of large bronchopleural fistulas and this has been our experience as well. Endotracheal Tube Obstruction Subtotal occlusion of the endotracheal tube or airway may result in refractory hypercapnea. An abrupt rise in PaCO2 during HFOV in an otherwise stable patient should be considered an obstructing or narrowed endotracheal tube until proven otherwise. If this is suspected, a suction catheter should be passed immediately to ensure patency of the endotracheal tube. Urgent bronchoscopy can then be performed either during HFOV or with a brief interruption of HFOV during manual bagging to visually inspect the airway. We have seen cases of endotracheal tubes obstructed by blood clots or mucus causing a “ball” valve effect that allowed passage of a suction catheter but would not allow exhalation to occur. This diagnosis can only be made by bronchoscopy. A sudden increase in ∆P (with no change in mPaw) may indicate a mainstem intubation or increase in airway resistance. A drop in mPaw (with no significant change in ∆P) may indicate a new or worsening airleak (10). 5. Sudden Stoppage of the HFOV Another aspect of HFOV is that there is no backup conventional ventilation mode. The oscillator may suddenly stop if decompression of the circuit occurs. This is usually because a pressureregulating mushroom valve or other external connection has become loose or disconnected. For this reason, it is imperative that a resuscitation bag with PEEP valve and attached oxygen source is always positioned at the head of the patient for immediate use. Should the oscillator suddenly stop, the patient should be immediately disconnected from the circuit and manually bag-ventilated while trouble shooting the ventilator. 6. Weaning From High Frequency Oscillatory Ventilation The principle goal of using HFOV in treating patients with ARDS is to achieve a nontoxic FiO2 (<60%) while minimizing ventilator induced lung injury. When patients respond with improved oxygenation, the first weaning maneuver therefore is to reduce the FiO2 before any reduction is considered in mPaw. We attempt reduction of FiO2 to 40% with a target SpO2 > 90% before attempting reductions in mPaw. If the patient can maintain a SpO2 > 90% on FiO2 40% we will than start a gradual reduction of mPaw (e.g. decrease 2 – 3 cmH2O q 4 – 6 hours as tolerated). If the SpO2 decreases during mPaw reduction, we will try to resume the previous mPaw that was able to maintain a SpO2 > 90% on FiO2 40%. It is important not to decrease mPaw too rapidly in an attempt to get the patient off HFOV. If the lung derecruits and desaturation occurs, it can take many hours to regain the lost volume. In our experience, attempting reductions in mPaw while the required FiO2 is still at 50 – 60 % will more often result in desaturations requiring re-institution of higher mPaw settings. Once a mPaw of 20 – 24 cmH2O has been achieved while maintaining an FiO2 of 40%, the patient can be switched back to a trial of conventional ventilation. We typically set up the conventional ventilator to achieve a mPaw of 20 +/- 2 cmH2O by using pressure control mode (PCV) with peak pressure set to achieve a delivered tidal volume of 6 – 8 ml/kg predicted body weight and inspiratory plateau pressure < 30 – 35 cmH2O. In determining the appropriate peak pressure setting it is important to titrate pressure based on delivered inspiratory tidal volume. For this reason we favor using ventilators that display both delivered and exhaled tidal volumes in the pressure control mode. Titration of pressure control based on exhaled tidal volume alone in our experience often leads to overestimation of the peak pressure required – especially in patients with cuff leaks or bronchopleural fistulas.