part0011

Without mechanical ventilation, patients with severe respiratory failure will undoubtedly die. Positive pressure ventilation can reduce shunt, improve gas exchange, and take over the work of breathing until the patient has recovered. That doesn't mean, however, that the ventilator can do anything to reverse the underlying condition or disease process that has led to respiratory failure.

Ventilator-induced lung injury (VILI) has been recognized as a necessary evil since the advent of modern critical care medicine, and to altogether eliminate the risk of any lung injury from the ventilator is not realistic. That said, much VILI is actually an unnecessary evil, since it occurs in the pursuit of "normal" gas exchange or "optimized" physiologic parameters. In cases of severe respiratory failure, the risk of VILI is high and the potential for rewards is small—it isn't reasonable to injure the patient's lungs in order to increase the PaO2 from 65 to 95, when 65 is sufficient to maintain life. Focusing on doing the minimum intervention necessary to support the patient is much more likely to be helpful in the long run.

Do what's necessary, not whatever is possible. Chasing the ideal of "normal" gas exchange will inevitably lead to VILI and unnecessary therapeutic interventions, all of which carry very real (and unwanted) side effects. With severe respiratory failure in particular, the twin objectives are to sustain the patient and minimize the risk of further injury. Sustaining the patient is obviously the more important objective, and there will be times when very high ventilator pressures are necessary to achieve it; however, anything that exposes the patient to real or potential harm should be justifiable .

As an aside, this can be the most difficult part of taking care of critically ill patients. We have all been taught what's "normal," and we all face the temptation to do things in order to bring things [lab values, physiologic measurements, vital signs] back into these ranges.

We use clinical studies and guidelines as a framework for therapy, but what works for one patient may not necessarily work for another. Additionally, the volume of evidence for the critical care of the most severely injured or ill patients is scant. Therefore, it takes a willingness to try different things and an ability to admit when a particular therapy isn't working. In these cases, protocols and clinical pathways can be harmful in that they can constrain physicians from trying a new approach to the problem.

Trying a different approach may be necessary. More often than not, however, the patient is adequately sustained with his current ventilator settings but the clinicians are tempted to change course in order to improve the numbers. This has the potential for harm without much benefit and should be avoided. Any modifications should be done either to lower the risk of injury or if the current settings are not providing an acceptable degree of life support. Keep in mind that the medical literature is full of therapies that improve oxygenation, ventilation, and vital signs. Very few of these have actually translated into better patient outcomes.

Patients with severe respiratory failure are in it for the long haul. This means that the chances of improvement in a few days are low and that the need for at least some mechanical ventilator support for several weeks or months is quite high. Couple this with the sedation requirements and relative immobility that accompanies endotracheal intubation, and it's obvious that the sooner the patient has a tracheostomy, the sooner he can begin some degree of mobilization and rehabilitation. A tracheostomy is associated with less sedation, more patient comfort, better mobilization, and fewer days on the ventilator when compared with the endotracheal tube. Do it as soon as it's safe.

Most patients with respiratory failure, even severe ARDS, will eventually recover. Those who survive ARDS will have near-normal lung function after six to twelve months. Even people with cardiopulmonary or neurologic disease who ultimately require a long-term tracheostomy can have an acceptable quality of life. Declaring a patient "ventilator-dependent" or saying that he has "no chance of recovery" after one or two weeks in the ICU may be premature or even wrong. Unbridled optimism isn't appropriate, but neither is pessimistic nihilism.

Some conditions are not survivable. Some conditions are survivable, and even have the potential for some recovery, but will leave the patient with significant disability and the need for partial or full ventilatory support. Lastly, some conditions are survivable and will require a prolonged period of critical care and ventilatory support, but with a chance at a full recovery to independence. Obviously, nothing is guaranteed, but clinicians caring for patients with respiratory failure should be able to discern which scenario is most likely and present this to the patient and his family.

Once a treatment plan is decided upon, it is imperative for the clinician to maintain a positive outlook. The patient and his family will be looking for encouragement and guidance, especially when there's a setback or a run of bad days. Throughout the course, the most important thing is open and honest communication. There are times when a shift to palliative care or hospice is appropriate—failure to recover, development of a new and severe complication, or if the patient is unwilling to continue a therapy with a small chance of success that is associated with significant discomfort or an unacceptable quality of life. In these situations, providing the patient and his family with a peaceful, comfortable death is a vital function of the clinician. There may be other times, however, when a setback is temporary and reversible, albeit discouraging (for example, development of pneumonia that requires going back to full ventilator support until it's adequately treated). Here, the clinician should encourage the patient and continue to focus on the ultimate goal of therapy, which is recovery to an acceptable quality of life.