ASA October 2003 Newsletter

When considering delivery of an “anesthetic” in the intensive care unit (ICU), anesthesiologists are commonly called to the ICU to help manage airways, to assist with procedures done at the bedside and to help manage uncommon conditions such as status asthmaticus. Before discussing the management of such patients, it is best to begin with an overview of current practice in critical care medicine.

ICUs arose for several reasons: the polio epidemics of the early 1950s, the development of coronary care units to take advantage of newer technologies for resuscitating patients with cardiac arrest and the desire of anesthesiologists to care for complicated cardiac and neurosurgical patients in close proximity to operating rooms (O.R.s). Having spent several hours with a difficult patient, anesthesiologists wanted their patients cared for in a location close to the O.R.s with better nursing care and using the technologies and therapies that anesthesiologists had been using in the O.R.

Over the last five decades, modern ICUs have diverged significantly; increasingly, they are not in close proximity to O.R.s, and a wide spectrum of physicians and health care providers deliver care. Only 1 percent to 2 percent of anesthesiologists in the United States are board-certified in critical care medicine and spend a significant amount of their time in the ICU. Though anesthesiologists who work in an O.R. environment occasionally provide care for patients in ICUs, they may not be familiar with some of the new technologies such as continuous venovenous hemofiltration and hemodialysis, ventilator strategies, nutrition support, etc.

Equally surprising to many of us would be the degree to which patients are “anesthetized” while they are in the ICU. It would not be uncommon in an ICU to have a ventilated patient on a propofol drip at 50–100 mcg/kg per minute along with an infusion of a neuromuscular blocking agent (NMBA) that rendered the patient chemically paralyzed. They are more anesthetized than some of the patients in the O.R.! In such a patient, one could do almost any surgical procedure without noticing much of a change in hemodynamics or evidence of autonomic hyperactivity.

Despite these observations, the majority of intensivists are not familiar with the anesthetic techniques, drugs and therapies that we utilize daily to manage patients in the O.R. In fact we in anesthesiology have learned much over the years when we have attempted to take with impunity therapies that we utilize in the O.R. to the ICU. For example nitrous oxide was once advocated as a wonderful drug to sedate patients in the ICU. When patients were so sedated for several hours to days, they developed megaloblastic anemia related to interference of nitrous oxide with DNA synthesis. Similarly etomidate, because of its “safety” profile, was advocated for sedation in the ICU, which led to the discovery of its interference with adrenal metabolism. Other adverse effects were identified when morphine (or other drugs, e.g., midazolam) was administered for prolonged periods of time, especially in patients with renal failure, and then discontinued. The sedative, analgesic and respiratory depression properties of the morphine lasted much longer than anticipated, secondary to active metabolites that had accumulated.

The use of an NMBA for an extended period of time (greater than 24 hours) in an ICU patient led to the discovery of what is now known as acute quadriplegic myopathy syndrome (AQMS). Patients who have an NMBA (either an aminosteroidal compound or a benzylisoquinolinium drug) stopped after an infusion of the drug for greater than 24 hours may remain completely flaccid and remain so for not just several hours but for days, if not weeks. The exact mechanism is unknown, but it is felt that approximately 5 percent to 10 percent of patients in the ICU who receive an NMBA blocker for extended periods of time are at risk for developing this syndrome. There also are concerns about not only prolonged effects but also drug withdrawal when sedative and analgesic agents are discontinued.

When we have taken drugs from the O.R. to the ICU, we have learned some of the dangers of the medications we commonly use in the O.R. when given for extended periods of time.

Returning to the issue of delivering an anesthetic in the ICU, let us first begin with airway management. Many patients who must be intubated outside the operating room are intubated because they have experienced a cardiac or respiratory arrest. In these circumstances, medications are infrequently given to manage the airway as patients are already unconscious and flaccid. Approximately 20 percent to 30 percent of the time, however, patients are severely ill, quite combative and must be pharmacologically managed to facilitate endotracheal intubation. In these circumstances, the drugs most commonly used are either midazolam, fentanyl or both; if the patient is very hemodynamically compromised, etomidate can be given. In 10 percent of patients who require pharmacologic intervention, NMBAs are administered. Succinylcholine is often used, though rocuronium is increasingly being given in this circumstance. If patients have hepatic or renal insufficiency, atracurium or cisatracurium could be tried. Probably more important than familiarity with medications in these scenarios is the ability to effectively manage the airway, providing mask ventilation if necessary, and direct laryngoscopy and endotracheal intubation. So when going to the ICU, an anesthesiologist should bring the appropriate drugs along with the appropriate airway management equipment.

Anesthesiologists also are called to the bedside to assist with tracheostomies or tracheotomies, for dressing changes in patients with burns and for invasive procedures such as transesophageal echocardiography or thoracentesis or for pleurodesis. As when anesthetizing patients in the O.R., certain information must be collected and equipment must be available prior to providing anesthesia in this environment. It can never be assumed that a patient’s status is nothing-by-mouth (NPO). Even NPO patients may have a full stomach either from a gastrointestinal bleed or if they are being fed enterally. Therefore if anesthesia services are going to be given for a bedside procedure, one must consider whether the patient has a full stomach, and if need be, an endotracheal tube should be placed using rapid sequence intubation for some of these procedures. In such a circumstance, perhaps the procedure should be performed in the O.R.

Though monitored anesthesia care can be provided for many bedside procedures, some patients (very agitated or in pediatric patients) may require a general anesthetic. In these circumstances, a total intravenous anesthetic is most commonly provided, and propofol with an opioid or an opioid and a benzodiazepine are administered. Many patients are already receiving a benzodiazepine and opioid or propofol infusion, in which case, the patient may already be intubated and ventilated. In this scenario, there is probably little for the anesthesiologist to do. If the patient is not intubated and is receiving low-dose opioid or midazolam, however, increasing the dose of either or both of these medications should be considered or adding a medication such as propofol. The disadvantage of using any of these medications as a bolus dose in a potentially hemodynamically compromised patient in the ICU is the hypotension and decreased cardiac output commonly seen in anywhere from 30 percent to 50 percent of patients. Though etomidate may be the better choice, an even better choice could be the use of ketamine along with a benzodiazepine in order to provide not only better hemodynamic stability but also better airway protection and pulmonary function.

With respect to the latter, on rare occasions patients with severe status asthmaticus will require the services of an anesthesiologist. In this circumstance, positive pressure ventilation is not necessarily ideal because of intrinsic positive end expiratory pressure, i.e., the stacking of breaths that may make ventilation and perfusion even more difficult. The use of an inhaled anesthetic agent, most commonly isoflurane, may be considered. An anesthesia machine is brought from the O.R. to the patient’s bedside. Depending on the ventilator or the anesthesia machine, it may be adequate to ventilate the patient, or the ICU ventilator may need to be adjusted to accommodate the addition of an anesthetic vaporizer. Provision has to be made for scavenging the anesthetic agent and for monitoring the patient periodically. Unlike in O.R.s, the anesthesia care provider does not stay at the bedside, though the nursing ratio must be one-to-one, and the ICU nurses must be educated and advised on what to do given different scenarios.

Conclusion
Anesthesiologists helped to develop modern ICUs but, because of changing practice patterns, have a decreased presence in the ICU. On occasion they will be called to the ICU to provide anesthesia care and need to be familiar with the drawbacks of providing this care and have a system in place to assure patient safety, adequate equipment to manage and facilitate airway control and appropriate pharmacologic agents to provide anesthetic care in the ICU. In doing so, we not only improve patient care, we provide an important venue for our ICU colleagues who are not anesthesiologists to become aware of the expertise that we can provide in assisting them in managing patients.

Michael J. Murray, M.D., Ph.D., is Professor and Chair, Department of Anesthesiology, Mayo Clinic, Jacksonville, Florida.

return to top