At 3 am, you respond to a nursing home for a 57 year old in severe respiratory distress. On arrival, the nurse presents you with a form entitled “Physician Orders for Life Sustaining Treatment” (POLST). As you are reviewing this, the patient goes into cardiac arrest. The POLST form clearly spells out that this patient would allow CPAP for respiratory distress, would allow IV fluids, but would not allow chest compressions, medications, or ventilatory support. It does not say “DNR” in big letters across the top or “Approved by Maine EMS” anywhere on the document. Can you respect this document?

As the medical community does a better job of respecting patient autonomy, and as more individuals are aware of the concept of end-of-life care, medical providers, patients, attorneys, and advocacy groups are working together to improve patients’ end-of-life care. Decisions and documentation of those decisions is being formalized. There is a gap in belief and action about end-of-life decisions; about 70% of people believe end of life planning is important, only about a third of Americans have a living will (36 percent) and 37 percent have identified a healthcare proxy or drawn up a durable power of attorney for healthcare. However, this is an important topic for most Americans.

Historically, the medical community and EMS have been suspicions of the concept of DNR. Working from the principle of “First do no harm,” there has long been an assumption that maximizing quantity of life was the greatest benefit to a patient; quality of life was felt to be of secondary importance and patient autonomy during life threatening events was occasionally given nothing more than lip service. In addition, there was (and persists, to a limited degree) paranoia that patient family members might have evil motives for descalating care. When the concept of DNR was introduced to EMS, therefore, it was with the caveat that the default behavior was resuscitation and only the most rigorous, state EMS approved, “i” dotted and “t” crossed DNR forms were to be grudgingly respected.

This attitude has evolved dramatically over the last decade, although this past summer’s discussion of “death panels” makes it clear that there is still a lot of education needed. Patient autonomy and consciousness about quality of life often being more important than quantity of life has driven many to do a better job discussing what needs to be done during emergency situations, particularly when the patient has lost decision making capacity. During this transition phase, there have been some excellent and some not so good attempts to clarify patient wishes. Some are absolutely clear while others vary between containing so little information as to be worthless and being so detailed that clinicians can’t find specific information about, say, cardiac arrest. Standardization of language and documentation has generally improved the usefulness of these documents. Nonetheless, EMS is still in the catch-up phase and transitioning from a mindset of “first resuscitate” to “first follow the patient’s wishes” is difficult.

Maine EMS, through the statewide protocols, has addressed the issue of valid DNR orders. Page 1 of the Gray section of the 2008 Protocols is “Do Not Resuscitate.” Item “I” clarifies when to start resuscitation. Section “II” discusses when not to start resuscitation. Section IIC addresses DNRs and reads as follows (my comments are inserted between the “*{}*” brackets):

When a physician Do Not Resuscitate (DNR) order is presented in one of three forms:

1. EMS DNR orders from other state EMS/DNR programs. If the order or device (e.g., plastic bracelet, jewelry, or card) appear to be in effect, and understandable to the crew, follow the order’s specific instructions. If there are no specific instructions beyond “DNR”, follow Maine EMS Comfort Care/ DNR Guidelines.

*{Since we have so many out-of-staters, it’s important to respect their wishes}*

2. Non-EMS DNR Orders - A written DNR order executed by a patient’s personal physician should be honored if it is understandable to the crew and if it is dated within 1 (one) year. Follow the order as written. If it is non-specific as to care to provide or withhold, follow the MEMS Comfort Care/DNR guidelines.

*{This is the key. If a DNR order executed by the patient’s physician is understandable and is dated within 1 (one) year, you honor it. It can be folded into origami. It can be written on a stone tablet. It can be red, blue, green, white, pink, or any color. It does not have to say “Approved by Maine EMS.” As long as there is a DNR order, executed by the patient’s physician and dated within ine year, follow the order as written. Note also that these orders don’t have to be limited to DNR. They can outline comfort care, allowable interventions, and other areas of impact on EMS. If there are conflicts with the protocols, contact medical control.}*

3. Maine EMS Comfort Care / DNR Program - A Maine EMS Comfort Care/ DNR order does not have an expiration date. Once activated, it remains in effect until the patient or someone acting on their behalf as described and authorized on the Comfort Care/DNR form cancels it. (Note: Although no longer distributed by Maine EMS, extant DNR/Comfort Care “orange” forms, wallet cards and plastic bracelets remain valid.)

One additional really important point is that the patient is the decision maker. If the patient has decision making capacity, then the patient’s wishes overrules anything written, anything the healthcare POA says, any other source of information. Be respectful of these other documents and decision makers, but the patient has the final say.

So, going back to the original scenario, what do you do? The POLST form, if its intention is clear and is dated within the past year, is a valid DNR under Gray IIC2. You respect it and follow the orders. If there is a treatment that conflicts with protocols, talk to medical control. But remember, respecting the patients’ wishes is critical, especially during the highly stressful times in which a DNR would be considered.

It’s worth your time to look at www.polst.org for more information.

Take care, do good, and remember, we help people achieve quality and dignity in life and and in death.

References:

Brandon E. “Planning Your End-Of-Life Care: Why filling out the paperwork isn’t enough. US News and World Report. September 14,2009. Accessed from http://www.usnews.com/money/personal-finance/retirement/articles/2009/09/14/planning-your-end-of-life-care.html 10/20/09.

www.polst.org Accessed 10/20/09.

We’re emphasizing thoughtful paramedicine in our paramedic program with the idea of nurturing excellent clinical judgment and so in the first week of class I presented one of my favorite topics, evidence based medicine and using the literature in EMS. The primary focus of this seminar is on understanding the basic principles of evidence-based medicine (ask a question, query the literature, critique the literature, and integrate the evidence with your experience into your practice of EMS). We also discuss some basic statistics (the only number you need to know is “5”) and describe the various types of manuscripts published in the EMS literature.

As part of our work to develop critical thinking skills, we are also introducing Enquiry Based Learning (EBL) and, more specifically, Problem Based Learning (PBL) into our curriculum. Although this practice is common in other fields of medical education, it’s pretty new in EMS and so today I ran a demonstration PBL for the class. For those of you who aren’t familiar with PBLs, our approach is basically to develop a clinical scenario broken into lots of little sections and after a small group reads each section, the participants come up with a list of things they don’t know about the content of that section, ask questions, assign the questions to the participants, and then the group separates, and each person researches his or her question. The group then comes back together and presents the answers to the questions. Highly effective and a great way to not only learn how to know what you don’t know but also how to learn the things you realize you don’t know. Since this was a demonstration, we ran through the scenario and then I answered the questions as if I were all of the participants (one of the advantages of writing the case is you know all the answers).

So anyway. The case was a cardiac arrest and one of the sections described the induction of therapeutic hypothermia (TH) upon arrival in the emergency department. The group realized it didn’t know much about TH and so that became one of the questions. When I subsequently answered the question, the discussion then arose about prehospital induction and, since it was an opportunity to do some teaching, we talked about that as well.

Rapidly approaching me was a “wow” moment. I’m talking about the current state of the prehospital TH induction literature and what it takes to get TH done in an ambulance when one of the students looks at me and says…

”If the evidence supports starting it before 8 hours, where is the evidence that starting it even sooner is better?”

OK, it’s probably pretty obvious to anyone who has taken care of a post-resuscitation patient in the back of an ambulance that you’re kind of, you know, BUSY and one more task load is painful to consider if it doesn’t actually make a difference. But “Where’s the evidence?!?”

Now, when I went through my paramedic class in ’93-94, we would never have asked that question. First of all, what we were learning was from the national standard curriculum and was therefore clearly the gospel of EMS and second, and perhaps more importantly, we never really considered that you could question the source of what you were being taught and that you could wonder if an intervention had actually been shown to make a difference.

But this is exactly what we’re trying to foster in our class and frankly in our state. EMS has advanced to the point that we need to, as a practice of medicine, really start figuring out what it is that we don’t know. Many of our practices are not based on evidence from any studies, let alone EMS studies. The true EMS research literature is thin (but rapidly growing).

Some EMS providers rely on their protocols to absolutely guide care; unfortunately, the complexity of our assessment and intervention tools and skills are now approaching the complexity of our patients and our ability to really, really hurt people with what we do has never been higher. Protocols alone do not provide the “protocol medic” with the ability to really care for these patients. For example. knowing that furosemide does not provide an immediate benefit in acute CHF, that it is only useful in the 50% of patients in acute CHF who are actually total body fluid overloaded, and that it dramatically increases mortality in the two big mimics of CHF (COPD exacerbation and Pneumonia / Pulmonary Sepsis), the thinking medic will elect not to push this medication until it is absolutely clear that the patient is in cardiogenic pulmonary edema and is total body fluid overloaded. The “protocol medic” however may say “Hmm, my ‘Respiratory Distress’ protocol says I can give lasix nitro, CPAP, Duonebs, and methylprednisolone to this patient. I think I’ll do all of them since I’m not sure what’s going on” with little benefit to anyone from the furosemide and the potential for significant harm to many patients.

Conversely, we also have the ability to do great good for our most ill patients. Being able to diagnose an ST elevation MI (STEMI) in the field, pre-activate the cath lab, and get the patient into cardiac catheterization sooner does reduce mortality. Being able to provide non-invasive ventilation to patients with acute cardiogenic pulmonary edema (CHF) does improve outcomes. But we wouldn’t have these tools if a dedicated (rabidly fanatical?) cadre of field personnel, physicians, and researches hadn’t questioned the basic principles of traditional EMS techniques.

Getting back to the “wow” moment, it was really great to have a student look at me as I talked about exciting potential visions of the future and jerk me back to earth with “Where’s the evidence?” We do need to look to the future but we also need to make sure that what we do lets us perform our fundamental life stabilizing skills in ways that we know make a difference without burying us in care that may not make a difference at all. Therefore, to all of you providers out there who ask every day “Can we do this better?” I give you my utmost support and appreciation. To those answering that question, we, and even more importantly, our patients, owe you an immeasurable debt.

Stay safe, ask questions, and remember, we help people.

Steve,
Thanks for an excellent question.  We’ve all been there.  62 year old male, COPD history, respiratory distress with wheezing, sats in the high 80’s, we give him an albuterol neb (powered by oxygen at 8 LPM to really make it bubble) and the patient suddenly goes apneic and dies and it’s all our fault….
Except that doesn’t happen.  The theory of the hypoxic drive mediated sudden hyperoxic death is a work of intellectual beauty, eloquence, and unfortunately, not actually an accurate description of anything that happens in the real world.
To give this whole discussion context, I’m going to make a simple statement that underlies all airway management: “Hypoxia kills, hypercarbia happens.”  In other words, if your carbon dioxide levels are high (to certain limits), the resulting increased acid levels of the blood (respiratory acidosis) are not overly harmful and the body can compensate for them.  On the other hand, if your oxygen levels are too low, you die.  Each organ has a different minimum oxygen level it tolerates, but the heart and brain are highly sensitive to low oxygen levels.
So again:  Hypoxia kills.  Hypercarbia happens.  Sick COPD patients die first and foremost of hypoxia, and while, as we will discuss, sudden increases in hypercarbia may contribute to sudden cardiac arrest, they are not the result of oxygen induced apnea.  Instead, by believing that the only way these people will die is if they go apneic from the oxygen, we completely miss that the respiratory failure is what kills them and forget that assisted mechanical ventilation and not just oxygen is the treatment of choice for respiratory failure.

Now, the short answer to the question is that the fact that people act based on the myth of the “hypoxic drive oxygen induced apnea” is actually far more lethal than the oxygen.  Patients with COPD (or chronic asthma) have a whole body oxygen deficit at baseline and are essentially in compensated respiratory distress.  Now worsen their pulmonary function and they go into decompensating and ultimately decompensated respiratory failure and die of hypoxia.  We have 3 (common) ways in EMS to support patients in decompensating respiratory failure:  face mask ventilation, CPAP / BiPAP, and invasive (intubation, LMA ®, King ®, Cobra PLA ®, etc, etc.) mechanical ventilation.  If someone didn’t have COPD, we wouldn’t withhold oxygen.  Why do we insist on withholding oxygen in ill patients who at baseline are hypoxic?  The power and fear of the myth are the dangers to these patients, not appropriate oxygenation.  If you were looking for an answer just for yourself, you can stop reading now.  However, if you have to defend this position, then please keep reading for a little support.

Let’s start with a discussion of why we breathe.  The short answer is that carbon dioxide, oxygen, and muscle stretch receptors in our chest drive respiration.

About 85% of our drive to breathe comes from carbon dioxide receptors that “measure” (through a number of mechanisms) carbon dioxide levels in the blood.   The receptors are in the brainstem and measures blood acid level (which is related to the carbon dioxide level).  After 24 hours of elevated carbon dioxide levels (“chronic ventilatory failure”), the body’s compensation system (using something called bicarbonate) changes the pH (a measure of acid level) at the carbon dioxide receptor and (insert mysterious action at which we can only guess at the wonders of the human body here) either decreases carbon dioxide sensitivity or, like a thermostat, increases the CO2 level necessary to drive respiration.  Either way, the CO2 receptors in these chronically hypercarbic patients still appear to have some function in the real world (which is where we take care of our patients).

About 10 - 15% of our drive to breathe comes from oxygen receptors located in the carotid arteries and the aortic arch.  Normally, your oxygen pressure (a way of measuring oxygen levels) in your arteries (PaO2) is 80-100 mm Hg (millimeters mercury, a common pressure measurement).  For most people, when the PaO2 drops to 60 mm Hg or below, there is a drive to breathe.  However, there are a number of factors including CO2 levels, acid levels, and insert even more mysterious actions at which we can only guess at the wonders of the human body here, which can change this level.  In other words, in people with COPD, this response to breathe may be triggered at higher oxygen levels (PaO2 above 60 mm Hg).  At a PaO2 above 170 mmHg, this trigger theoretically turns off since you are no longer hypoxic (however, read the Aubier article for a great study which shows just how smart our bodies really are).

Finally, there are a variety of conscious and unconscious triggers that also drive you to breathe.  These include stretch receptors in your chest wall, a concern that you’re going to die if you don’t keep breathing, and people telling you to breathe.

So, what happens when a COPD patient becomes ill, either from a COPD exacerbation or from other causes?  Regardless of cause, they almost always become even more systemically hypoxic (low oxygen levels throughout the body).  Every organ is under stress, particularly the heart and the brain.  They are initially in compensated respiratory failure; that is, they are skating on the edge but no longer truly capable of adequately oxygenating.  Their PaO2 drops and they breathe harder to try to oxygenate; they also blow off excess carbon dioxide and change blood flow in their lungs to maximize perfusion (blood flow) with oxygenation (open, working alveoli).  When we apply oxygen, a number of really complex events occur that actually increase the PaCO2 (level of carbon dioxide in arterial blood).   For an excellent description of this, see http://home.pacbell.net/whitnack/The_Death_of_the_Hypoxic_Drive_Theory.htm.  However, it is important to understand that the rise in PaCO2 is mostly from normal physiologic response to restoration of alveolar oxygen levels, increased deadspace, changes in pulmonary capillary blood flow, and decreased CO2 elimination; only a small amount in a minority of individuals can be blamed on decreased minute volume.  This sudden increase in CO2 can be harmful (possibly even fatal) but remember, we have ways of helping people get rid of CO2; those methods are known as assisted or mechanical ventilation. But before we starting blaming ourselves for not expecting this sudden non-apneic increase in CO2, remember that these patients started out in respiratory failure and while high flow oxygen may improve their oxygenation status, respiratory failure is much bigger than just oxygen levels and that little bit of oxygen you’ve added doesn’t fix the respiratory failure.

At this point, we have a patient who probably feels a little better because his heart and brain are finally getting oxygen.  But he’s still fatigued, the underlying pathology still exits, he’s still in decompensating or decompensated respiratory failure and his CO2 levels are increasing because of blood and gas redistribution in the lungs (but not from decreased minute ventilation or apnea).  The patient might stop breathing because of the fatigue, the bronchoconstriction, etc, (but not from normal oxygen levels).  Therefore, now that we’ve oxygenated the patient, we need to intervene in the respiratory failure.  Non-invasive ventilation is a key first step to keep the patient from fulfilling the mythical legacy of hyperoxic apneic death.  It’s all about cause and effect.  Your oxygen won’t be the cause of the patient’s apnea and death.  Your failure to intervene for the respiratory failure and rapidly increasing CO2 levels just might be.

Clinical bottom line?  Patients in respiratory distress, whether CO2 retainers (COPD patients) or not, need oxygen.   Patients in compensated or decompensated respiratory failure, whether CO2 retainers (COPD patients) or not, need ventilatory support.  Patients with COPD who are otherwise healthy will, with very few exceptions, tolerate supplemental oxygen without any effects and the rare group who does have an effect don’t go apneic but rather breathe a bit more shallow than before and become a bit sleepier.  If your COPD patient does go apneic when you apply oxygen and you didn’t expect it, then you probably failed to recognize just how sick they really were.

Be safe, play well with others and remember: we help people.

References:

Aubier M.   Effects of the administration of O2 on ventilation and blood gases in patients with chronic obstructive pulmonary disease during acute respiratory failure. Am Rev Respir Dis 1980;122:747-754.
Nice study that demonstrated that even when you blew through the upper limits of the hypoxic drive (PaO2 of 225 mm Hg), the respiratory muscles did just fine and while ther was an intial decrease in minute ventilation by 15 minutes there was essentially no difference.

Caruana-Montaldo  B, Gleeson K, Zwillich GW.  The control of breathing in clinical practice. Chest 2000;117:205-225.
A good review of how and why we breathe normally and in various disease states.

Crossley DJ, McGuire GP, Barrow PM, Houston PL. Influence of inspired oxygen concentration on deadspace, respiratory drive, and PaCO2 in intubated patients with chronic obstructive pulmonary disease. Crit Care Med 1997; 25(9): 1522-1526.
A great study using intubated but spontaneously breathing COPD patients that demonstrated both that apnea did not occur and provided further confirmation of the mechanism of increased CO2 which has nothing to do with apnea or descreaed minute ventilation.

Hoyt JW.  Debunking myths of chronic obstructive pulmonary disease.  Crit Care Med 1997;25:1450-1451.
A good summary discussion of the Crossley manuscript.

Nunn JF: The pulmonary circulation. In: Nunn’s Applied Respiratory Physiology. Fourth Edition. Nunn JF (Ed). Oxford, UK, Butterworth-Heinemann, 1993, p 145.
A nice summary of the way pulmonary circulation actually works.

http://www.rtmagazine.com/issues/articles/2000-02_17.asp (accessed 2/27/9).
A nice summary of the issue plus a really interesting discussion of the subset of patients who do decrease (minimally) their respiratory rate and level of consciousness at an FiO2 of .3 (academically interesting).

http://home.pacbell.net/whitnack/The_Death_of_the_Hypoxic_Drive_Theory.htm (Accessed 2/27/9)
Another interesting summary of the issue from the perspective of a respiratory therapist.

We had an interesting hypothetical question today that I thought would be a good jump off for the blog.

You are presented with a hypothetical 16yo female hyperventilating with carpal spasm. After a thorough assessment you determine that there is no detectable medical stimulus for the rapid breathing and it is most likely due to an anxiety issue (previous diagnosis). A provider on scene suggests you should shut the O2 off on the NRB mask to promote rebreathing of CO2. I was hoping you might weigh in with your thoughts.

Thanks,

DB

Dr. Busko Responds:

Hyperventilation is “excessive ventilation; specifically: excessive rate and depth of respiration leading to abnormal loss of carbon dioxide from the blood; called also overventilation” (1).  Hyperventilation can be done on purpose.  For example, when a skin diver hyperventilates, he lowers the carbon dioxide levels in his blood and decreases the drive to breathe (remember that most of us are triggered to breathe when the carbon dioxide levels in our blood rise).  It can also happen unintentionally and is called “psychogenic hyperventilation,” “hyperventilation syndrome,” “behavioral breathlessness,” or “psychogenic dyspnea.”  The hyperventilation results from the underlying behavioral condition.

When carbon dioxide levels get low the blood is less acidic (more alkalotic).  This makes the salts in the blood and the cells (the electrolytes) move in and out of the cells in ways they usually wouldn’t.  These “electrolyte shifts” cause many of the symptoms the patient has.  In particular, the calcium shift causes the carpopedal (hand and foot) spasms that are very painful.  The patient can also have general weakness and numbness and tingling (low blood phosphorus), leg cramps (low blood potassium), central nervous system symptoms including syncope and seizure (decreased blood flow to the brain), chest pain with EKG changes (from many electrolyte changes), and wheezing (bronchospasm caused by low blood carbon dioxide levels).  Although the risk of death from psychogenic hyperventilation is low, it can have major impacts on the body.

Even more importantly, it is important to figure out whether the patient has hyperventilation (as defined above) versus tachypnea (fast breathing) or hyperpnea (deep breathing), both of which are signs of underlying disease.  The major difference is that in hyperventilation the minute ventilation (how much volume moves in and out of the lungs each minute) is more than the body needs; in tachypnea and hyperpnea, the breathing rate and/or volume is increased because the body’s metabolic demand has increased.  Can you really rule out a pulmonary embolus as the cause to the patient’s rapid or deep respiratory rate?  Pneumonia?  Early congestive heart failure?  Carbon monoxide poisoning?  Sepsis?  Diabetic ketoacidosis?  Spontaneous pneumothorax?  In all these cases, the respiratory rate is increased because the body needs more oxygen.  And a patient with a history of hyperventilation syndrome can develop all of these diseases.

In emergency medicine and EMS, our job is not so much to diagnose what a patient has, but rather to make sure that nothing bad is going on.  If we happen to diagnose something benign (not bad) along the way, great, but most importantly, we need to be sure that we’re not missing something bad.  Since psychogenic hyperventilation is a diagnosis of exclusion (that is, all that’s left after you’ve made sure nothing else is going on), you need to be really sure it’s what you’re dealing with so that you don’t miss anything else important or life threatening.

The theory of having a patient rebreathe carbon dioxide, whether from a brown paper bag or a non-rebreather without additional oxygen flowing in, is that the rebreathed carbon dioxide increases the level of carbon dioxide in the blood and reverses the problems caused by the respiratory alkalosis (low blood acid levels caused by low carbon dioxide).  There has never been a study to show that this treatment for acute hyperventilation syndrome actually works.  What is much more concerning is that there are many deaths from acute MI (2), asthma (3), DKA (4), pulmonary embolus (5), and other organic diseases (2) that resulted from the patient being diagnosed with “acute hyperventilation syndrome “ and not being worked-up or treated for their real underlying disease.  These reports apply to emergency department and EMS patients.  Furthermore, a major contributor to these deaths is hypoxia.  Patients with underlying organic disease have increased metabolic needs.  When carbon dioxide rebreathing is used, they actually get lower oxygen than is in room air (21%) because very little new oxygen is being added to the paper bag or non-rebreather.  The technique, quite literally, suffocates the patient and studies from as long ago as 1989 demonstrated rebreathing to be a dangerous treatment technique (6).  Acute hyperventilation syndrome may be uncomfortable but it is not life threatening.  Misdiagnosising a life-threatening condition by calling it “just hyperventilation” may be.

No matter what the cause, the assessment and treatment of the patient who is breathing rapidly includes:
-Protect yourself (lots of environmental toxins and atmospheric conditions cause rapid ventilation)
-Perform good initial assessment and resuscitation interventions (“resusassesment” as needed
-Try to determine the underlying cause but remember that many of the tools used in the emergency department to rule out bad stuff are not available in the field (and therefore you can never rule out all the bad stuff)
-Prevent hypoxia.  They may not all need high flow O2 but no one needs to be put into a hypoxic environment either (e.g. rebreathing techniques)
-Treat things that you find (asthma, STEMI, etc…)
-Do not use rebreathing techniques
-Coach the patient if there appears to be an underlying psychological component
-Never blow these patients off as having “just hyperventilation”

Be safe, play well with others and remember, we help people.

Additional Resources
Kern B, Rosh AJ.  Hyperventilation Syndrome.  Emedicine.com.  Retrieved February 02, 2009, from emedicine.com website: http://emedicine.medscape.com/article/807277-overview
http://emedicine.medscape.com/article/807277-diagnosis
http://emedicine.medscape.com/article/807277-treatment

References:
(1)    hyperventilation. (n.d.). Merriam-Webster’s Medical Dictionary. Retrieved February 02, 2009, from Dictionary.com website: http://dictionary.reference.com/browse/hyperventilation
(2)    Saisch SGN, Wessely S, Gardner WN.  Patients with acute hyperventilation presenting to an inner-city emergency department.  Chest 110(4);1996:952-57.
(3)    Gardner WN, Bass C, Moxham J. Recurrent hyperventilation tetany due to mild asthma. Respir Med 1992; 86:349-51
(4)    Treasure RAR, Fowler PBS, Millington HT, et al. Misdiagnosis of diabetic ketoacidosis as hyperventilation syndrome. BMJ 1987; 294:630
(5)    Kern B, Rosh AJ.  Hyperventilation Syndrome.  Emedicine.com.  Retrieved February 02, 2009, from emedicine.com website: http://emedicine.medscape.com/article/907277-treatment
(6)    Callahan M. Hypoxic hazards of traditional paper bag rebreathing in hyperventilating patients. Ann Emerg Med 1989; 18:622-28