Keith Lamb, RRT-ACCS
Adult Critical Care Supervisor
UnityPoint Health System
Des Moines, IA
Joe Hylton, RRT-NPS, CPFT
Brady Scott, MS, RRT-ACCS
Clinical Education Coordinator and Assistant Professor
Department of Respiratory Care
Rush University Medical Center
Keith D. Lamb, RRT-ACCS
I was recently asked to present a lecture to critical care nurses about hypoxia and hypoxemia. While preparing I thought back over the last few years to recap in my mind what we have studied, and what we have learned, about these topics. At the end of my preparation, I was once again reminded that the paradigm regarding the support of patients with these gas exchange derangements has shifted from “make the numbers look textbook” to “please don’t hurt the patient.”
This is true in almost all that we do in critical care. Terms like “permissive” hypercarbia, acidemia, hypoxemia, and hypotension have become commonplace. These management ideologies have come from a now vast abundance of science and data suggesting that hurting the patient is very easy to do when attempting to make the numbers look like they do in the textbooks.
The first term that many of us heard was permissive hypercarbia. This strategy became popular when it was first realized that using larger tidal volumes was harmful. When tidal volumes were dropped, carbon dioxide levels went up. No big deal. We know that higher levels of carbon dioxide are fairly benign. Hypercarbia is well tolerated in the majority of patients. We know that using tidal volumes in excess of 6-8 ml/kg PBW is harmful to patients when they have sick lungs. Permissive hypercarbia allows us to ventilate with safer volumes. Volutrama, barotrauma, and resultant biotrauma are avoided. This improves outcomes.
Permissive acidemia has also gained traction. Like hypercarbia, respiratory acidemia is well tolerated, and as a matter of fact, is self-corrected fairly quickly in patients with adequate renal function.
Metabolic acidemia is a little trickier. Treatment with sodium bicarbonate is unpopular, primarily due to its rapid disassociation into water and carbon dioxide, and to a lack of evidence that outcomes are improved. Using the respiratory system to “over correct” for metabolic acidemia is equally unpopular due to a clear disadvantage to cerebral perfusion when PaCO2 is significantly below normal levels. This same premise is also responsible for a decrease in the use of hyperventilation in acute brain injury.
On the flip side, enzyme function in the face of severe acidemia is sub-optimal. Patients receiving vasopressor therapy for shock refractory to fluid bolus do not tolerate this acidemia well, and hemodynamically unstable patients can sometimes benefit from exogenous sodium bicarbonate infusion.
You can see why each clinical scenario should be approached individually, and although moderate acidemia can be tolerated in most ICU patients, there are some who cannot be managed this way.
Low oxygen levels can also be well tolerated, particularly when the alternatives are known to be injurious. Risk and benefit should always be evaluated. High levels of PEEP can be counter-productive in patients with refractory shock and can contribute to acute hypotension.
Recruitment maneuvers may be injurious if not implemented in the correct patient population, and the use of high FiO2 levels has been proven to worsen outcomes by causing free radical production and interference with basic cellular function, resulting in multi system organ failure.
Finally, exogenous nitric oxide administration has not been proven to improve outcomes in refractory hypoxemia even though plasma oxygen levels are acutely improved in most. Conversely, inhaled nitric oxide has shown a trend to a higher incidence of acute renal failure.
The risks and potential benefits appreciated with the use of extreme positioning, high frequency oscillatory ventilation, and other modes of ventilation such as APRV and BiVENT are not well understood, although it has recently been suggested that prone positioning affords an improvement in outcomes when used with patients with very sick lungs. Modalities that solely improve oxygen levels most likely do not positively affect outcomes and mortality. Those modalities that improve oxygen levels AND subsequently allow the clinician to use less injurious ventilator settings (i.e., significantly lower FiO2 levels and lower distending pressures) may in fact make a difference.
It is believed by many that the improvement in outcomes seen in patients who are placed in the prone position are actually not from increasing oxygenation but rather from an improvement in the mechanical characteristics of the lungs themselves, which in turn leads to a lower incidence of secondary lung injury.
Once again it has been discovered that an attempt to correct numbers to their textbook levels can lead to disastrous outcomes. It has been learned that attempting to restore blood pressure to its previously normal levels in some hypotensive patients can have deleterious effects. This is very true in trauma patients. Pre-hospital placement of large bore intravenous access and the infusion of large amounts of crystalloid can “pop” compensating blood clots, dilute blood volume with non-oxygen carrying fluid, and cause acute coagulopathies that worsen hemorrhage.
Permissive hypotension in an attempt to allow the body’s compensatory mechanisms to work, and avoiding the associated negative sequelae of abruptly increasing blood pressure with large volumes of fluid, has gained popularity during recent experiences in combat. These same principles hold true in urban combat and other trauma as well.
There are very few things we can do to “fix” our critically ill and injured patients. Alternatively, there are MANY things we can do to cause them harm. Our mission is to avoid falling into the trap of “fixing” numbers before first asking ourselves if we are in fact hurting our patients.
Mary Ann Couture, MS, RRT-ACCS
Hartford Hospital, Hartford, CT
Brain death, a primary neurological disease, is determined by the irreversible loss of function of the brain and brainstem. Apnea testing is the final test in determining brain death. For diagnosis, an apnea test result is positive if there is at least a 20 millimeters of mercury (mm Hg) increase in arterial carbon dioxide tension (PaCO2) from a baseline normal PaCO2. Absent respiratory movements have to be observed for the entire period of the test.
Associative testing problems may include not reaching the target PaCO2 goal, requiring test repetition and hypoxemia. Complications can include pneumothorax, arrhythmia, hypertension, hypotension, hypoxemia, and irreversible cardiac arrest. These problems can be exacerbated when pre-existing complications are present, and it is possible that the test could compromise viability of organs destined for transplantation due to the hypoxemia or cardiovascular instability.1According to Arbor, additional problems could be due to metabolic and hemodynamic instability related to the brain injury pathophysiology.2
Brain death standards when defining criteria for irreversible coma were first presented in 1968 by the Ad Hoc Committee of the Harvard Medical School.3 Later, in 1981, the President’s Commission produced the Uniform Determination of Death Act, which led to federal legislation adopted by all 50 states.4
The American Academy of Neurology (AAN) published evidence-based practice parameters for brain death in 1995,5 with an update in 2010.6 Both reports established current standardized neurological testing and described the apnea test in detail. However, the apnea test is not universally accepted by all countries. The standard criteria to assess neurological symptoms include clinical absence of neurological function and deep coma that is assessed prior to the apnea test. After a single clinical evaluation, the apnea test is given as the final test. The AAN guidelines were meant to standardize the procedure, although there is room for other tests such as an EEG and cerebral angiogram when the apnea test cannot be performed.
The following is a partial list of guidelines for the apnea test after coma, or when absences of brainstem reflexes are established:6
A PubMed search from 1995 revealed a few studies that tried alternative approaches to the apnea test. Melano et al. randomized 200 tests to either the AAN guidelines or an indwelling catheter with CO2. They experienced a high level of complications in the augmented gas group, although hypoxemia was higher in the control group. In both groups combined, they noted 21% complications.7
Lang’s group assessed 43 patients in two randomized groups, adding CO2 at 1 LPM in one of the groups via a catheter within the ventilator circuit tubing near the patient’s inspiratory airway for 1-2 minutes while connected to the ventilator. It was reported that the danger with this process was to significantly overshoot the target PaCO2. Lang did suggest trying augmented CO2, with hypoventilation via the ventilator to increase PaCO2, along with end tidal monitoring and serial ABG.8
In 2003, Sharpe and colleagues introduced the concept of maintaining ventilatory support at a very low frequency rate while giving 3% carbon dioxide and 97% oxygen (carbogen) from a compressed gas cylinder that was added to the oxygen supply line. They utilized the pre-test arterial pH with a formula to target an endpoint pH which, under normal circumstances, would drive the rise in PaCO2. They applied capnography to monitor that CO2 rise in 60 patients. The results appeared reliable, with an added benefit of not disconnecting the patient. There was no lung derecruitment and fewer associative complications, although the test took longer.9
A small randomized crossover study of 20 adults was done by Lévesque and colleagues. Ten minute tests were repeated in random order with three systems:
They found that the T-piece and CPAP were effective alternatives, with O2 best maintained with CPAP.10
The 2010 AAN guidelines still defend the use of CPAP as the only acceptable alternative procedure. However, carbogen has been regularly presented at the AARC Congress as an acceptable alternative method.
Since brain death is considered to be a neurological disease caused by severe head injury, aneurysm, or hemorrhage, it is usually a neurologist who makes the diagnosis. Occasionally, physicians see brain death from lack of oxygen and tissue loss (hypoxic-ischemic) brain insults that cause loss of brain function, but there are fewer patients where this occurs. Examples are cardiac arrest, drug overdose, or asphyxia injuries.
Brain death studies are low volume procedures that few hospital staff encounter. In many large hospitals this diagnosis is made around 25-30 times each year. While apnea testing is important it can also exacerbate brain or systemic injury due to hypoxemia or cardiovascular instability. Major factors and errors may be due to lack of pre-oxygenation and tests that are performed by different physicians with different techniques and different endpoints to the test.11,12
Before attempting a test, certain goals need to be met to try to maintain near normal body temperature and blood pressure, and to ensure adequate perfusion to all organs potentially destined for donation. As a general rule, a request to a family for organ donation cannot be made until brain death is diagnosed. So staff should treat all patients as if they were potential donors.
Since brain death can lead to hemodynamic instability, fluid and electrolyte imbalance causing metabolic instability, and challenges with oxygenation, complications can be responsible for a loss of potentially viable donor organs. Even body temperature de-stabilization may affect renal function. While hemodynamic stability is the responsibility of nursing care, the respiratory therapist can assist the team by ensuring respiratory standards are met prior to the neurological assessment.
Not all brain dead patients are eligible to become organ donors. Yet the possibility needs to be considered since associative testing problems related to ventilator disconnection and timing of the ABG with the apnea test may include not reaching the target PaCO2 goal, which would require test repetition and further exacerbate brain or systemic injury. A literature review from nine studies with a total of 608 patients showed frequency of complications as follows: hypotension 18%, hypoxemia 6%, arrhythmia 1%, bradycardia <1%, and cardiac arrest <1%.13
On any given day there are approximately 121,000 people waiting for transplants, with an active daily list of 79,000. These data can be found on the United Network for Organ Sharing website. The donor list for the entire year of 2013 was 14,256.14 Maintaining tissue and organ integrity can maximize the number of donations that a single patient can make. In order to provide care for potential donors, it is important to understand the pathophysiology of the disease progression and how it can affect the lungs and all other organs
Brain death physiology is usually characterized by two distinct hemodynamic phases: a hypertensive crisis with massive sympathetic discharge followed by neurogenic hypotension with an upload of pro inflammatory mediators. An ischemic brain and the initial high cerebral blood pressures may cause Cushing’s reflex, autonomic destabilization, and the associative cascade effect of a “catecholamine storm” release. This is then followed by hypotension requiring more fluid boluses and vasopressor and inotropic management, which can affect renal function. In other words, poor hemodynamic maintenance from the beginning will result in increased cytokine levels that will affect many organs, including the lungs.16,17
There is some belief that the early hypertensive crisis triggers the pro inflammatory response. This was explained by Avlonitis and colleagues in an animal model wherein brain death was induced in 56 rats that were randomized after either a balloon catheter induced brain death or a sham catheter.17 Some rats received treatment for the hypertensive crisis with boluses of an a-adrenergic antagonist, phentolamine, which eliminated the hypertension but had no role in preventing the hypotensive period. During the neurogenic hypotension phase, another group received continuous infusion of noradrenaline. The sham group had the highest cytokine levels, suggesting the brain death process triggers the inflammatory response. There was a significant decrease in cytokines in both of the groups that received the a-adrenergic antagonist in the early hypertensive crisis and the group that received noradrenaline in the hypotensive crisis. Furthermore, the pretreatment a–adrenergic group had better oxygenation and the lowest levels of cytokines found in the lungs.
Lung injury in the donor is the main reason for low lung utilization rates for lung transplants, so care must be maintained. After brain death determination, a heightened continuation in critical care management ensues, with a shift in goals. The aftercare may become even more rigorous and time-consuming. Plus, extra time is needed for meticulous matching to recipients and communication to the appropriate recipient parties.
For lung preservation, low tidal volumes to prevent ventilator-induced injury should be maintained to prevent further cytokine release. If lungs are to be harvested there may be additional tests ordered such as bronchoscopy and biopsies. Maintaining the highest PaO2/FiO2 (P/F) ratio at the lowest FiO2 possible is very important as some lung transplant facilities have a cut off point for the P/F ratio. Prevention of pneumonia and atelectasis can be achieved by implementing ventilator associated pneumonia bundles, frequent recruitment maneuvers, adequate PEEP, and additional pulmonary toilet, which may be ordered prophylactically.
An example of how low the lung harvest selection can become when management is poor was provided by Mascia et al. They looked at 34 patients from 15 ICUs in 13 hospitals in Italy. Of the 11 (32%) eligible lung donors, only two lung donations occurred. Of those potential lung donors, 5 (45%) had a P/F ratio of <300. The authors found that even though hemodynamic management was performed, the ventilatory management remained the same, with no escalation in derecruitment or prevention maneuvers once brain death was declaired.18
Until the patient is declared brain dead, he is at risk of death from the apnea procedure itself. This reflects upon the expertise of the caregiver, who must recognize and eliminate unstable clinical signs before and during the test. Another important reminder is that extra care is required from all team members to maintain tissue and organ function, especially if an organ donation is expected to occur. Lung injury prevention strategies are needed before and after death is established. Once declared, and if the patient is to be an organ donor, it becomes the caregiver’s responsibility to utilize all forms of respiratory care at an elevated level to prevent further damage, not only to the lungs but to all organs.
Specialty Practitioner of the Year: Use our online nomination form to nominate a fellow section member for our 2014 Adult Acute Care Specialty Practitioner of the Year award.
Recruit a New Member: Know an AARC member who could benefit from section membership? Direct them to online sign-up. It’s the easiest way to add section membership to their overall membership package.
Bulletin Deadlines: Winter Issue: December 1; Spring Issue: March 1; Summer Issue: June 1; Fall Issue: September 1.