This case report raises several issues on the difficult subject of DNR orders and in-hospital resuscitation, especially when the life threatening situation is not caused by expected complications (for example, due to acute deterioration of underlying diseases) but is related to rare and sometimes trifling causes. Furthermore, our case report also emphasises the dilemma of DNR orders in the presence of iatrogenic complications.
A recent review1
reported an incidence of in-hospital cardiopulmonary arrests of 1–5 arrests per 1000 patient admissions, and therefore resuscitation efforts are routinely performed in hospitals. Outcome can be determined by: (1) an immediate survival with a return of spontaneous circulation as direct success of the cardiopulmonary resuscitation manoeuvres; and (2) the more important survival to discharge as many patients die during the first 24 h after an initially successful resuscitation. The most commonly quoted outcome is the survival to discharge, which varies from 0–42%, with the most common range in the major studies of approximately 20%.
There are several conditions in different groups of patients that decrease the rate of survival to discharge to virtually nil in some cases.2
These are conditions such as sepsis on the day before resuscitation, metastatic cancer, dementia or a serum creatinine value at a cut-point of 1.5 mg/dl (130 μmol/l). Another important predictor of survival is the type of arrest, with highest rates of survival to discharge after primary respiratory arrest and lowest after pulseless electrical activity or asystole.3
On the other hand, improved survival rates after resuscitation are observed for witnessed cardiac arrests.
There are, of course, some patients with cardiopulmonary arrest in whom resuscitation should not be performed—that is, where a DNR order has been written. The basis for such decisions is given by the main ethical principles that comprise beneficence, non-maleficence, justice and autonomy.4
As these ethical principles also apply to resuscitation issues, DNR orders should be considered in patients who object to CPR due to comprehensible reasons or in patients in whom a CPR would be futile due to medical considerations. In a prospective cohort study on the factors associated with DNR orders, the writing of them was mainly influenced by the patients’ preferences followed by the probability of surviving the next 2 months.5
Another interesting finding of this study was the fact that only 52% of patients who preferred not to be resuscitated actually had a written DNR order. With respect to good clinical practice, therefore, it must become a front ranking task to improve that number urgently. Beyond that, awareness of the DNR orders in the clinical background also has to be notably improved. For example, in radiology departments the compliance with DNR orders is low.6
Twenty per cent of respondents to a questionnaire sent to hospital based radiology departments would resuscitate patients with DNR orders, and 38% had resuscitated patients with DNR orders in the past. Reasons for performing CPR in patients with DNR orders were, among others, unawareness of the DNR status. This was also the case in our patient with a written DNR order, where anaesthetic standby was ordered for the diagnostic CT scan and, even more controversially, where resuscitation was actually performed.
In several clinical situations, a pre-existing DNR order can be reconsidered and discussed. Following the main ethical principles mentioned before, this should be limited to situations where the decision for the DNR order is exclusively based on medical considerations—that is, where CRP is expected to be futile due to the underlying diseases and cardiopulmonary arrest occurs under special conditions (such as in the operating room, etc) with demonstrable improved rates of successful resuscitations. In fact, DNR orders have been routinely suspended in patients undergoing surgery during the intraoperative and immediate postoperative period.7–9
Another situation where discussions about overriding DNR orders occurs is in cardiopulmonary arrest during haemodialysis, as these arrests are witnessed and frequently triggered by iatrogenic causes.10
Iatrogenic complications occur as an adverse reaction to treatment or as a failure to treat. In hospitalised patients 14% of cardiac arrests follow an iatrogenic complication that are mainly medication errors and toxic effects (44%), or suboptimal response by physicians to clinical signs and symptoms (28%).11
Patients with an iatrogenic cardiac arrest were more likely to survive to discharge. There is no consensus in the literature about how to deal with iatrogenic cardiopulmonary arrests in DNR patients. Some authors state that they will perform resuscitation only if there is reason to believe that it may be consistent with the patient’s intention.10,12,13
Others feel that there are situations where resuscitating a patient in an iatrogenic arrest can be medically and ethically justified.12
In a survey with three different hypothetical case scenarios, physicians were unlikely to override a DNR order in patients suffering from a cardiac arrest originating from the underlying disease.13
In patients who arrest due to a complication of medical treatment, however, physicians were much more likely to resuscitate and even more physicians would resuscitate a patient whose arrest was caused by a physician’s error. A “limited aggressive therapy order” has been discussed for patients with a DNR order in situations with a likelihood of “higher success”, such as witnessed cardiopulmonary arrests with ventricular fibrillation or tachycardia as initial rhythm, an arrest in the operating room, or a cardiac arrest due to an easily identifiable iatrogenic cause.14
If, however, the DNR order is due to the patient’s declared intention and/or due to ethical considerations, the authors of this report clearly disclaim suspension of DNR orders irrespective of the current situation—for example, whether the arrest occurs during surgery, ICU stay, and haemodialysis, or even if the cause for the cardiopulmonary arrest is an iatrogenic one (always under the assumption that the medical action that had triggered the arrest was free of any malicious intention). In doing so, however, this by no means excludes taking the greatest efforts to correct for triggering causes and mistakes, but any such efforts have to be below the threshold of CPR—that is, intubation/mechanical ventilation (unless already preinstalled) and cardiac massage. In case the arrest is due to ventricular fibrillation and occurs in the operating room, an attempt at electrical defibrillation might be justified. In daily practice, however, suspension of DNR orders is often managed in a different (albeit less differentiated) way.
In our patient, cardiopulmonary arrest was not related to his advanced stage of comorbidities but brought on by an unexpected bagatelle—that is, a bolus aspiration of the gauze from the nasal tamponade. During CPR, the obstacle was easily removed, resulting in a successful resuscitation without sustained harm. Unexpected and/or easy to resolve conditions could scarcely be excluded as a potential cause of cardiopulmonary arrest, and thus may give rise to neglect by principle DNR orders due to medical considerations in nearly all—except some very specific—situations. Well understood, such considerations are independent of the circumstance whether the life threatening complication is iatrogenic in nature or not. With respect to similar cases, and knowing about successful resuscitations also in patients with apparently futile medical comorbidities, we would strongly encourage neglecting DNR orders that have been written exclusively due to medical considerations or, even more consistently, to desist from DNR orders if triggered by circumstances other than ethical considerations and/or the patient’s comprehensible will.
To sum up our personal recommendations concerning DNR orders, we would strongly encourage checking the patient’s intention concerning CPR on a routine basis—for example, at hospital admission. In case there are ethical considerations against CPR and/or the patient refuses possible CPR on a comprehensible basis, then a DNR order should be followed under any circumstances, independent of whether the arrest occurs during surgery or is due to an iatrogenic cause. By contrast, we would disclaim DNR orders that are triggered exclusively due to medical considerations—that is, in patients with advanced stage comorbidities in whom CRP is expected to be unsuccessful and/or might result in harm. We would definitely encourage the initiation of CPR in such patients, but would terminate such endeavours early if the patient was unresponsive.
In the case of our patient, the arrest was due to an iatrogenic cause which was easily resolved, and thus the CPR was successful and by no means regretted by the patient. Nevertheless, the patient still prefers having a DNR order for the future.
- In patients with an otherwise undisputed DNR order, ethical dilemmas can arise when the arrest is: (1) due to a bagatelle (that is, easy to resolve but not easy to detect); (2) iatrogenic in nature; and (3) occurring in a unmonitored area (for example, the ward), as compared with an arrest in a well monitored area (for example, ICU and operating room).
- Our personal recommendation in these situations would be to follow the DNR order strictly if based on the patient’s intentions, but to neglect the DNR order (and, thus, to start cardiopulmonary resuscitation) if based exclusively on medical considerations.