Fall 1998, Vol.5, No.2 The Slow Code: An Ethics Case Conference Jeff D. Hardin, M.D., Ph.D. Resident, Department of Medicine New York Presbyterian Hospital The following article contains excerpts from a recent ethics talk given in the Department of Medicine at Columbia-Presbyterian Medical Center. Dr. Hardin was one of eight senior residents to serve as a Jay I. Meltzer ethics fellow in the 1997-1998 academic year. This ongoing program is sponsored by the Vidda Foundation and the Columbia University Center for the Study of Society and Medicine, which is coordinated by Dr. Barron H. Lerner. Case Presentation L.J., a 54-year-old woman with end-stage ovarian cancer was admitted to a teaching hospital with sepsis. Throughout her hospitalization, the emergency department (ED), intensive care unit (ICU), and oncology teams all addressed the topic of a do not resuscitate (DNR) order with the patient and family members. However, L.J. still harbored hope that another treatment would become available and she wished that, in the event of cardiac arrest, ". . . everything be done." Her doctors felt frustrated and helpless in the face of her advanced disease, but promised to follow her wishes. In the middle of the night during the second week of admission, she experienced a cardiac arrest and a code was called. Resuscitation was initiated with bag-and-mask ventilation and chest compressions. Epinephrine and atropine were called for by the resident, but were secretly injected into the mattress instead of the patient's intravenous line. After one round of medications, the resident ordered a cessation of efforts and pronounced the patient dead. Introduction In hospital wards throughout the country, certain patients experiencing cardiac arrest are receiving partial, half-hearted attempts at cardiopulmonary resuscitation (CPR). So called "slow codes" are performed on patients who have preexisting poor prognoses, but have full resuscitation orders. Slow codes, also known as "partial", "show", "light blue", or "Hollywood" codes, are cardiopulmonary resuscitative efforts that involve a deliberate decision to not be aggressive.1 In these cases, there is often discord between the expectations of the hospital staff and those of the patient and his or her family or health care proxy. The slow code has become an unspoken rite of passage for many house officers. Factors that have fostered the use of the slow code include medical paternalism, miscommunication between hospital staff and patients or their surrogates, and medical futility. In addition, patients may have unrealistic expectations because medical dramas on television routinely depict miraculous resuscitations.2 Additionally, among physicians in teaching hospitals, there is often dissent among the ranks regarding code status. The house staff who are charged with the day to day burden of patient care and the messy reality of conducting a code may not agree with the attending physician's support for full code status. Worse, some attending physicians may avoid addressing the issue of DNR with tenuous patients and their families. In these instances, the house staff may feel that they are dumped upon and left to clean up the mess. Overworked and faced with what they perceive as medical futility, some house officers see the slow code as their only recourse. Despite the widespread practice of slow codes, many physician groups and professional organizations have taken positions against its practice. In 1983, the President's Commission on Ethical Problems spoke out on the subject: [P]artial resuscitation [involves a] less than full effort to resuscitate the patient . . . because the attending physician never made a clear decision or because it was thought important to placate or comfort family members or hospital staff. Success at resuscitation is rare enough when all efforts are expended, so such limited efforts are usually doomed from the start. Thus 'partial codes' become a dishonest effort that needs to be justified by reasons stronger than merely the providers' discomfort in discussing DNR decisions.3 Other groups also reject the slow code as a therapeutic option. The American College of Physicians' Ethics Manual states: "If DNR orders are not written, it is unethical for physicians and nurses to perform half-hearted resuscitation efforts (so-called 'slow codes')".4 Furthermore, the American Heart Association's Advanced Cardiac Life Support Manual, which is widely accepted as the standard of care when it comes to cardiopulmonary resuscitation, remarks: 'Show codes' or 'slow codes' appear to provide CPR while not actually doing so or while doing so in a way that is known to be ineffective. Slow codes, particularly when done to deceive the relatives and friends of a patient, compromise the ethical integrity of healthcare professionals and undermine the physician-patient or nurse-patient relationship. Such codes should not be done.5 Despite these strong objections to performing slow codes, evidence suggests that this type of code is frequently undertaken. An informal survey conducted by the author indicates that the majority of the Columbia-Presbyterian Medical Center housestaff has participated in some type of slow code, and almost all of these residents were comfortable with their participation.6 Another published study found that residents at Seattle area teaching hospitals classified 5 to 10% of patients on certain medical wards as "slow codes." 7 Cardiopulmonary Resuscitation The slow code is a relatively recent phenomenon that has grown out of advances in CPR techniques. Prior to the 1960's, CPR was rarely attempted and patients experiencing cardiac arrest were allowed to die without intervention or were subjected to the bloody and often futile technique of open-chest cardiac compression. The first published series of closed-chest compressions during in-hospital cardiac arrests launched an era of aggressive resuscitation as well as the unrealistic expectation that all patients experiencing cardiac arrest have a good chance of survival. 8The investigators in this early study reported 70% survival among patients who arrested while undergoing anesthesia. This high figure has never been reproduced. More recent inpatient statistics give an average of 30% immediate survival of cardiac arrest, with 15% survival to discharge9-12. Several studies show that patients with severe underlying illnesses (e.g., metastatic cancer, sepsis, pneumonia, or stroke) who experience arrest have less than a 1% chance for long-term survival9,13,14. Despite these grim odds, the majority (40-90%) of in-patients wish to undergo CPR in the event of cardiac arrest.15-17 Medical Futility The Oxford English Dictionary Second Edition defines futility as: "incapable of producing any result; failing utterly of the desired end through intrinsic defect; useless, ineffectual, vain."18 Proponents of medical futility describe it as having both quantitative and qualitative aspects.19 The probability of success (or failure) of a medical intervention embodies the quantitative element of futility; Schneiderman et al argue that if "in the last 100 cases a medical treatment has been useless, [physicians] should regard the treatment as futile."19 Qualitative futility is defined by the utility of the outcome. Although certain therapeutic measures may have a quantitative likelihood of being successful, the outcome may not benefit the patient in any meaningful sense. An example of this would be the renal dialysis of a patient who has severe, irreversible brain damage. In judging futility, physicians must distinguish between an effect, which is limited to some part of the patient's body, and a benefit, which appreciably improves the person as a whole. Treatment that fails to provide the latter, whether or not it achieves the former, is 'futile.'19 In the face of quantitative and especially qualitative futility, many authorities recommend that therapeutic limits be set on patients experiencing cardiac arrest. The American Heart Association's Advanced Cardiac Life Support Manual states: Medical futility justifies unilateral decisions by physicians to withhold or terminate resuscitation under the following circumstances: Appropriate basic life support (BLS) and advanced life support (ALS) have already been attempted without success. No physiological benefit from BLS and ALS can be expected because a patient's vital functions are deteriorating despite maximal therapy. No survivors after CPR have been reported under the given circumstances in well-designed studies.20 Other authors suggest a more assertive position during discussions of advanced directives with patients who have poor prognoses. Physicians discussing resuscitation with such patients, for example, may propose to the patient or his or her surrogate that DNR is the most medically sound decision: "In cases in which the physician has firmly decided that a 'no-code' order is the proper course, it usually works out better for him to explain to family members why resuscitation will not be attempted than to ask them whether or not they wanted it attempted." 21 This approach may unburden the family members from the responsibility of making the decision to let go of their loved one. Nevertheless, given that current legal and ethical standards place decision-making in the hands of the patient and/or family, most physicians are unwilling to take such a unilateral approach. Conclusion Several factors have led to an increase in the frequency of slow codes. An aging population that has more comorbid conditions and receives more health care has an increased likelihood of suffering a cardiac arrest in the hospital. Additionally, patient expectations of the efficacy of medical interventions such as CPR are higher than ever.2 Therefore, although educational efforts about DNR orders and medical futility are ongoing, it is likely that codes will continue to be performed on patients in whom such measures will be futile. Despite the widespread practice and acceptance of the slow code by medical staffs, its use undermines communication and trust between patients and physicians: "At best, it is a waste of time and a failure to face the reality and hard decision-making; at worst it is an ethical fraud."22 Most ethicists agree that the time has come to end the practice of the slow code. Rather, physicians should discuss advanced directives early and frankly and, when appropriate, should educate patients and their families about both quantitative and qualitative medical futility. Discussion Physician 1: [The criticism has been given that] the reason that we have slow codes is that DNR isn't addressed in a timely manner...We [often] watch somebody's demise over several days. And the family [is] there and we meet with them serially...[But] we constantly encounter their refusal or denial about resuscitation efforts. At every step of the next aggressive intervention we approach them and say, "Do you really want us to do this?" and get a "yes"...I really don't know if there is a timely way. Physician 2: Generally the slow code on the ward takes place after DNR has been discussed. The arrest resident doesn't make the decision to [have] a slow code. Usually the resident or intern who takes care of the patient and knows the situation gives us some clue that this [should] be a slow code. When it comes to a slow code, most of the time [DNR] has been discussed. Physician 3: In some situations it is very difficult for the family to take responsibility even if [DNR status] is what they wish for their family member. Physician 4: We are talking about situations of futility, where patients who have been on maximal support [and for whom resuscitation] is either not going to work or if it works it will only work for an hour and then [he or she] will arrest again. Why do these codes have to be carried out? Why can't we say, "This is a futile situation. We have good statistics behind us and simply refuse to do it." We are too often put in the situation of having to carry out a futile code. Physician 5: I personally would be against...[making] an inaccurate disclosure in the chart. I think it is wrong to inject a drug into the [patient's] mattress and say it has been given. As far as I am concerned, you can use your clinical judgment in carrying out a resuscitation...And if you feel that a code can be terminated in two minutes...[after] we have tried [several interventions without success], nobody is going to be sued or taken to court. You can say, "We did the best we could. We tried what we could." Physician 6: There seems to be a discrepancy between the [quantitative definition of] futility versus what our visceral sensation of futility is. It is tough to quantify our clinical decision whether or not something is futile. The concern is that there is a slippery slope. It is unethical to do something that constitutes an undue burden or torture to the patient when there is no benefit. Physician 7: We have to separate the [arrest team] and the physicians in charge who know the patient and are making the decisions. The [arrest] team needs to have the prerogative to make the decision when there has been a state of cardiovascular unresponsiveness. This can be ascertained in 5 minutes or 25 minutes. Comment by Dr. Barron H. Lerner, Angelica Berrie Gold Foundation Assistant Professor of Medicine (in the Center for the Study of Society and Medicine): Slow codes are one of the most challenging ethical issues that medical students confront when they first begin work on the wards. During the first two years of medical school, students learn that patients and their families have the autonomy to make clinical decisions; deliberate deception of such persons is, by definition, unethical. However, as Dr. Hardin has shown, the daily realities of the inpatient wards introduce students to an alternative perspective. House officers, many of whom also learned about the slow code in medical school, now argue that they are being forced to provide futile and even inhumane treatment to patients with no meaningful chance of recovery. In this scenario, the slow code becomes the lesser of two evilsan unfortunate strategy for dealing with an intolerable situation. Medical students witnessing slow codes are thus torn in two directions. On the one hand, they greatly respect the views of their housestaff mentors; on the other hand, they feel uncomfortable with the acts of deception they are witnessing. It is thus essential that medical schools and residency training programs address the issue of slow codes in an open manner. By giving doctors-in-training better instruction in end-of-life decision making and palliative care and providing them with emotional support services, better communication with patients and families can be achieved. In such a setting, the temptation to perform slow codes will hopefully abate. References 1. Gazelle G. The slow code: should anyone rush to its defense? N Engl J Med 1998;338; 467-469. 2. Diem SJ, Lantos JD, Tulsky JA. Cardiopulmonary resuscitation on television. N Engl J. Med 1996; 334: 1578-82. 3. President's Commission for the Study of Ethical Problems in Biomedical and Behavioral Research. Deciding to forego life-sustaining treatment: a report on the ethical, medical, and legal issues in treatment decisions. Washington, DC GPO 1983; 89-90. 4. American College of Physicians. Ethics Manual, 3rd Edition. Ann Int Med 1992; 117; 947. 5. American Heart Association. Advanced Cardiac Life Support Manual. 1997; 16-17. 6. Hardin JD. Unpublished survey of Presbyterian Hospital medical residents 1998. 7. Neher JO. The "slow code":a hidden conflict. J Fam Pract 1988;27:429-430. 8. Kouwenhouven WB, Ing D, Jude JR, Knickerbocker GG. Closed chest cardiac message. JAMA 1960; 173:1064-1067. 9. Taffet GE, Teasdale TA, Luchi RJ. In-hospital cardiopulmonary resuscitation. JAMA 1988; 260: 2069-72. 10. DeBard ML. Cardiopulmonary resuscitation: analysis of six-years experience and review of the literature. Ann Emerg Med 1981;10:408-16. 11. McGrath, RB. In-house cardiopulmonary resuscitationafter a quarter of a century. Ann Emerg Med 1987;16:1365-1368. 12. Markert RJ and Saklayen, MG. Cardiopulmonary resuscitation on television. N Engl J Med 1996; 335:1605. 13. Bedell SE, Delbanco TL, Cook EF, Epstein, FH.Survival after cardiopulmonary resuscitation in the hospital. N Engl J Med 1983; 309;569-576. 14. Moss AH. Informing the patient about cardiopulmonary resuscitation. J Gen Intern Med 1989;4:349-355. 15. Schonwetter RS. Educating the elderly: cardiopulmonary resuscitation decisions before and after intervention. J Am Geriatr Soc 1991;39:372-377. 16. Finucane TE, Shumway JM, Powers RL, D'Alessandri RM. Planning with elderly outpatients for contingencies of severe illness: a survey and a clinical trial. J Gen Intern Med 1988;3:322-325. 17. Wagner A. Cardiopulmonary resuscitation in the aged. A prospective study. N Engl J Med 1984; 310;1129-1130. 18. The Oxford English Dictionary Second Edition. 19. Schneiderman LJ. Medical futility: its meaning and ethical implications. Ann Int Med 1990; 112:949-954. 20. American Heart Association. Advanced Cardiac Life Support Manual 1997; 16-22. 21. Spencer SS. Sounding board: "Code" or "no code":a nonlegal opinion. N Engl J Med 1979;300:139-140. 22. Goldenring J. "Code" or "no code" decisions [letter]. N Engl J Med 1979;300:1058. copyright ©, Columbia-Presbyterian Medical Center [Go to start of Document]