Critical Care Medicine - List
http://www.pitt.edu/~crippen/

David Ryon resurrects CPR guidelines.

David Ryon:

Furthermore, an order to spare the patient (and staff) from this procedure currently requires a consentual agreement from the patient or family. Often, in such circumstances, the family is already distraught over the catastrophic events leading to the patient's moribund condition. Ethically, it is difficult to pose this "decision not to resuscitate" to families in situations where we already know that the procedure will be entirely ineffective in reversing the problem. We often see the (already stressed) family then struggle to reach this "decision". I feel that this is a common reason that physicians do not seek a DNR order. It is easier to perform these "last rites" than to belabor the family with further decisions. This approach leaves the staff with the unpleasant and ethically difficult duty of having to pump on the dead person's chest for several minutes. Though the task may seem simple, it is repulsive to the nurse who has been caring for the patient all day and perhaps earlier in the patient's course. It requires involvement by a physician, respiratory therapist, and another couple nurses who could be doing more productive things.

The argument continues that, under certain circumstances, the order not to perform CPR is not comparable to an order to forgo life-sustaining procedures. The example that immediately comes to mind is the case where a patient has an escalating pressor requirement in the absence of reversible causes, and is finally unable to maintain his/her blood pressure. Chest compressions are never successful in recapturing spontaneous circulation in this scenario. Therefore, chest compressions would not be a life-sustaining procedure in this circumstance. Defining a set of medical indications for CPR would legitimize decisions to order "no CPR" in settings where it would not alter outcome. One ought to be able to write the order in this setting without the mandate of raising the issue with surrogates, for the reasons noted above.

This issue has been raised in the critical care literature and I would very much appreciate input from this group. I seek input and advice from lurking lawyers, ethicists and interested physicians. My plan is to identify patient populations who never immediately survive CPR in the hospital based on our own data and the literature. The initial policy can then be drafted and reviewed with our ethics committee (followed by executive committee and medical staff). Meanwhile the identified patient population can then be evaluated prospectively prior to the institution of the policy.

I can see that I will be accused of getting on a "slippery slope", but I seek input from physicians everywhere to assess whether other institutions are initiating a similar approach to the problem. Consider the case of the patient who has been fully evaluated but remains refractory to indicated pressor support (say, due to a massive MI). He is not a candidate for balloon pump or other more aggressive intervention. He has been on high doses of epinephrine and dopamine infusions.

What do you do as the pressure falls?

1. Approach the family and tell them of the situation. Tell them all interventions are not working and that CPR will be futile. Ask them to sign papers for DNR.

2. Inform the family of the situation and tell them interventions are not working. Mention that you will not be putting the patient though chest compressions since it will be useless. Note their response and if they appear to nod, then write the "no CPR" order in the chart and document your discussion.

3. Inform the family that the situation is grave. When the patient finally loses his BP, you cancel the "code" the moment it is called.

4. Inform the family that the situation is grave, but that you are doing all you can. You allow a moment for them to raise any objections. There are none. When the code is called, you follow ACLS protocol for several minutes until it is clear that the patient indeed has not restored spontanious circulation.

5. Inform the family that interventions are not working and that the patient will die shortly. Allow them into the room to be with him. There is no discussion of CPR. Write a "no CPR" order in the chart, documenting the situation.

Linda Oakes:

Re: your questions on CPR and DNR issues....I am not sure I understand exactly your narrative part but can understand the basic dilemma nurses, physicians, etc are put in when there is no plan for this patient... But I will answer by commenting on each of the approaches you mentioned at the end.

1. We do approach the family's...usually takes much more than one conversation as to how all the interventions are not working and that CPR or some parts of it will likely be futile. But we do not have the family member sign a paper. We feel that puts additional and unnecessary burden on them--"putting the pen to the paper" is hard.

2. Our approach is more like this one. But we don't just mention it we get their concurrence verbally. This conversation is to be witnessed by a nurse, chaplain, or social worker in order for them to put a note in the chart as well as the physician writing a DNR order--specific references as to what was discusses and agreed upon. So we need more than the appearance of a nod. Wouldn't recommend that, could be misunderstanding what the family thinks is the plan, leading to more pain and confusion at the time of arrest.

3. Never this for all the reasons you have outlined in your narrative...ie moral anguish of those who have to initiate a non-code. Not fair when attending physician is not in the building and longer delays in such a decision.

4. same as c.

5. doesn't fit our philosophy of including the family and patient when possible in the treatment plan. just a fantasy in minds of us to think the family will not wonder and ask why we will not be doing something...

For our patients/families including them in discussions on how CPR is usually successful, but health care workers still make the decision with their agreement. FOr the times it takes longer for family and health care team to come to agreement...we accept it is families that have to live with this much longer. SO we give them a bit more time...that has been discussed on this list multiple times.

AndyEgol:

I often wonder how we got into the situation we are now in. Most of the time I just can't understand it. CPR was devised as a methodology to reverse a reversible situation. To this end it was both noble and extremely useful and beneficial to a whole host of survivors who had reversible disease. It was never intended nor has it ever been useful to reverse irreversible situations. I am somewhat confused as to why or how CPR can be made analagous to other therapies such as IABP or LVAD. While I understand and agree that they are similar, they are treated anything but similar by hospitals and physicians. With IABP's we have developed a set of indications/contraindications over many years. With this experience we now have a better idea of who will benefit and who will not benefit from an IABP. While this "list" and "experience" may not be perfect, and mistakes of omission and commission are made, for the most part we are pretty good about when to use this technology.

Because we are pretty good at it, we do something different here we don't do with CPR. In patients who don't meet criteria for IABP insertion, WE DON'T ASK THEM IF THEY WANT IT INSERTED! Novel approach. However, with CPR, somewhere along the line we got screwed up and began asking people if they wanted us to do CPR on them even though they had no worldly clue as to what CPR is, when is it beneficial and when it is not. This is truly amazing. We do the same thing with ventilators and pressors and a whole host of other treatments that are somehow "different" from IABP, LVAD's, etc. By asking permission to do or not to do something immediatly takes the choice away from the only provider who understand the indications, contraindications, etc. This is truly amazing. In many papers on the ethics of resuscitation, as well as the position papers on life support it is clearly implied that physicians are under no obligation to provide treatments they feel are not beneficial. Yet we do so on a daily basis! Why? The autonomy issues of the 70's and 80's clearly defined the rights of patients to refuse therapy.

In the 90's we have gone beyond the autonomy of refusal and have extended autonomy to include the demand for therapy whether it is useful or not. Are there limits to autonomy? I certainly hope so. If not, we will be subjected to continously provide therapy where we feel there is no hope. Many of the autonomy issues of the 70's and 80's were decided in the courts. It is likely that the limits of autonomy will be defined there also. The issue of futility may need to be defined by the health care team using the best available tools for decision making available and treating it no differently than decisions to operate/not operate, give chemotherapy/not give chemotherapy etc. I believe the families should have these issues discussed with them, but the ultimate decision regarding indications, contraindications, utility, or benefit of proposed treatments are ones that can only be made by the personnel possessing the knowledge and experience regarding these issues.

Currently, fear of litigation and ethical confusion over the limits of autonomy have precluded us from being effective in this area. Policies may not resolve these issues which are fundamentally emotional, personal, fear based and confusing. Only time will tell. The literature is pretty clear on survivability of CPR in a variety of situations. Most of it is poor with discharge rates of 10%. When you eliminate the patients resuscitated post sudden death the true rate becimes truly abysmal. We don't do a consensus every time we come up with indications/containdications for procedures or therapies. We should not have to with CPR either. CPR is a procedure like all others and should be offered when it is indicated and not offered when it is not.

We make decisions regarding outcome all the time we just don't think about it in either statistical or absolute terms. We determine that operative intervention in an 85 year old previously healthy patient with acute MR and cardiogenic shock is or is not a candidate. Why do we make a decision for or against surgery? Does the team rely on a consensus to arrive at a decision? Would a variety of teams arrive at different decisions? We will never arrive at a decision for CPR that is absolute, just as we will never determine which is more important a 1% or 10% risk of being wrong. We will spend a lot of time and effort trying to do the right thing and having far greater expectations that we will come up with the right answers moreso than with any other therapeutic modality. This is I believe an unrealistic goal.

Howard Klions:

You raise some interesting questions regarding how we should handle inappropriate CPR. This is something we all struggle with daily. The problem as I see it often comes down to the definition of benefit. The number of cases where we can say with absolute certainly that CPR will be unsuccessful is small compared to the number of patients at risk. And our prognostic abilities are not as good as we would like to think they are when applied to individual patients. There are some patients who we probably could invoke futility and withhold CPR against a families wishes, such as the patient with progressive hypoxia in spite of full vent support or progressive shock on industrial strength pressors. I suspect many of us would not even consider CPR in these patients.

However, the problem usually comes down to a question of quality of life rather than effectiveness of CPR. Who is to make the decision whether a 1% chance of survival is worth taking, let alone a 10% chance? We can't predict which individual patients will survive and we usually can't predict their functional status if they do survive other than to say it certainly won't be better than their previous baseline. Statistics are usually meaningless when applied to the individual patient. Another problem is how are we to decide how much pain or disability it takes to be worse than death? One person's hell may be an acceptable quality of life for someone else.

The solution as I see it is in the hands of the patients family physician. Patients need to understand their medical condition and the consequences of "doing everything possible". The all too frequent scenario is a stat consult for a patient in extremis at 3AM and the sleepy family member on the phone doesn't know me from Adam and isn't about to trust me when I recommend comfort care. There is no excuse for the primary physician not having dealt with this in advance.

Another problem of course is when everything has been discussed in detail, but the family has unrealistic expectations and refuses to give up. I do think there is a roll for "futility" here, but until we have a better method of prognosis than currently available, the definition of futile is going to allude us. For futility to work, it has to be an accepted community standard, and that will require a reliable scientific method of determining outcome. I just don't think we are at a point where we can really develop a set of guidelines for CPR that even a simple majority of the physicians in a community can agree on, let alone getting the citizens of the community to accept them as well.

What we need is education, first of the primary care physicians, so that they may properly educate the public. And this will only scratch the surface of the problem.

Robert Levy:

I absolutely agree with Howard. I work in a small town in Oregon, so I am both the "family physician", as well as the" intensivist". Some of my partners are good about these discussions, and some are not. I strongly feel that I get a better decision from the patient, or the family, when the patient is not blue and gasping. It is hard to make an informed consent when you are hypoxic, when your loved one looks desparate, or it is 3 AM, and the family receives call from some unknown physician with unknown judgement . We need to encourage our primary care colleagues to address these problems at the appropriate time, as it provides better patient care.

David Ryon:

I agree, but the recent SUPPORT study (JAMA. Nov 22 1995;274:1591-8) suggested that this increased communication with the patient and family did not lead to significant changes in timing of DNR orders, days on mechanical ventilation, or number of days comatose before death. Furthermore, many preterminal events are not predicatable in time to discuss care preferences while the patient is well. I think there are larger problems out there:

1. Patients / families maintain some denial of the gravity of circumstances until they occur, perhaps as part of a natural psycological defence mechanism in dealing with a catastrophic diagnosis.

2. There is distrust in the physician's ability to prognosticate, as highlighted by recent commentary on CCM-L "what do you think of this" line started by FL.

3. There is some distrust in the physician's interests in making terminal recommendations. That is, with the advent of managed care, there is an economic conflict of interest in minimizing days of futile ICU care. The public is made aware of this through interest stories and novels focusing on this potential problem.

4. There are often problems interpreting patient's prior wishes when an acute event arrives. That is, most living wills do not outline limitations of therapy in situations that are thought to be clinically reversable.

5. There is often little incentive for a busy office practitioner to raise the delicate and time-consuming issue of DNR while the patient is healthy.

These observations are not meant to undercut Dr. Klion's thoughts, but merely point out how this complicated issue needs to be addressed from many angles.

Michael Darwin:

Howard's statements are wise and practical. But as I am often neither myself, here is what I would recommend:

1. Leaving aside the organ donor issue I would say that closed chest CPR (CC-CPR) in typical SCD should be discontinued if in-field, on-site defibrillation/definitive therapy is not successful.

2. CPR is not an effective modality in complicated sepsis, advanced neoplastic disease, or most cases of MSOF. In fact, in neoplastic disease it can be argued that sprinkling holy water on the patient has a much better chance of good outcome since this modality has NOT been proven worthless. I seem to remember one series of patients who experienced SCD in neoplastic diease (perhaps 10K patients in the study!) without a single CPR survival. IMHO it should not be administered to such patients any more than you would give cisplatin to a 1.5 sq meter, 32 kg cachectic CA patient with agonal gasping in end-stage breat cancer.

3. Closed chest CPR stinks. If you think a patient can really benefit from CPR, crack his or her chest and do open chest CPR (OC-CPR). If you don't think the patient can make it through the *minor* trauma of OC-CPR you are wasting your time. And it *is* minor trauma.

4. If you must use CC-CPR use high impulse, machine delivered CPR and don't go through this nonsense of long pauses between shocks (= MAJOR ischemic time) looking for NSR. Use a machine which delivers better CPR, puts your personnel out of the line of fire, and can shape the wave-form leading to much improved cardiac output and better outcome (do a literature search looking for work by Ornato, et al., if you doubt me). Or, call Michigan Instruments, Inc. in Grand Rapids, MI. And please, those "Thumper-haters" out there, save me thr trouble of your cries. I have used Thumpers for 20 years and there is no comparison between mechanical and manual CPR; the Thumper wins every time. With high-impulse CPR the Tumper leaves manual CPR in the dust by indices such as CO, EtCO2, pulse oximetery.

5. CC-CPR should continue to be used in cases of cold water drowing and hypothermia and in the young who arrest secondary to drug intoxication. But only when there is no one around to do OC-CPR or no high impulse Thumper around.

6. The chest excursion of 1.5 to 2" advocated by the AHA and Red Cross are all wet. Deeper deflections on down stroke are needed to optimise CO in most patients and they DO NOT break more ribs or crush hearts in my experience. Further, a Thumper with adjustable compression to relaxation ratios and other other "settable" parameters will help you optimise CPR for a given patient; as we all know, the response is variable with some patients being awake during CPR while others remain cyanotic and pulseless.

7. Objectify what you are doing. If you don't have good EtCO2s and can't get them with pressors, increased chest deflection, or whatever, STOP THE CODE. Disposable and fairly reliable EtCO2 sensors like the EasyCap are available for about $15.00 each. Every code cart should have one. If you are not getting CO2 out, forget it. Save the drugs, save the time, save the money. Short of not doing codes on hopelesss cases, STOPPING then before the bretyllium, the 10th defibrillation, and 45 minutes of everyone's costly and precious time and emotional reserve (for the typical 15 minute code) are just common sense. If you guys could objectify just how bad CC-CPR is you'd do a whole hell of a lot less of it in the first place, and even less still once you've started it.

The place to start in stopping CC-CPR may well be in excluding those cases where it is apparently absolutely contraindicated, and then getting quality, real-time feedback about its efficacy when you *do* use it. A lot of EtCO2 readings of 0.5% ought to tell you something. Like, maybe you are wasting your time and money.

Finally, if the family wants the contraindicated code, make 'em pay for it (I know, you can't do this). However, a return to OC-CPR which is vastly superior in terms of outcome would make the casual use of CC-CPR seem a hell of a lot less innocous and make the grandkids think about whether they want granny to have an emergent thoracotomy with a gloved-hand squeezing her dear old heart. Yeah, yeah, I know, I'll get hate mail on this one. But what I've said here has real merit.

Dick Burrows:

5. is the option I would choose. But I would tell them I am not going to jump on the chest. That the situation is not only grave - it is hopeless.

I can't help but feel that people have lost sight of common sense issues. I know that it is a trying time and that lawyers and all sorts of aggressive minority interest groups are stiring the pot but isn't this rather like telling the pilot of an aircraft or the captain of a ship that he cannot make any decisions in respect of who abandons ship first.

Do we have ANY authority? Or are we too scared to exercise any authority that we may have? Or are we scared of each other when we are called to give expert testimony against each other?

David Ryon:

Agree with Mike Darwin, though I'd have to admit that hacking open Granny to perform open chest CPR in order to spite a well-intentioned family is going a bit far.

The comments regarding clinical scenarios where CPR is not helpful are useful toward creating a protocol. Currently, in most US institutions that I am aware of, we must document discussion with the patient or surrogates whenever we write an order to withhold CPR. I am still looking for feedback from around the country and world to see how this is (or is not) handled. Let me repost this survey, please take a moment to respond / comment:

Consider the case of the patient who has been fully evaluated but remains refractory to indicated pressor support (say, due to a massive MI). He is not a candidate for balloon pump or other more aggressive intervention. He has been on high doses of epinephrine and dopamine infusions. What do you do as the pressure falls?

1. Approach the family and tell them of the situation. Tell them all interventions are not working and that CPR will be futile. Ask them to sign papers for DNR.

2. Inform the family of the situation and tell them interventions are not working. Mention that you will not be putting the patient though chest compressions since it will be useless. Note their response and if they appear to nod, then write the "no CPR" order in the chart and document your discussion.

3. Inform the family that the situation is grave. When the patient finally loses his BP, you cancel the "code" the moment it is called.

4. Inform the family that the situation is grave, but that you are doing all you can. You allow a moment for them to raise any objections. There are none. When the code is called, you follow ACLS protocol for several minutes until it is clear that the patient indeed has not restored spontanious circulation.

5. Inform the family that interventions are not working and that the patient will die shortly. Allow them into the room to be with him. There is no discussion of CPR. Write a "no CPR" order in the chart, documenting the situation.