A reply

We thank Professor Smith and Dr Poplett for their interest and encouraging comments on these guidelines. They propose simplification of the recommendations with a 2-option approach to managing patients with pre-existing Do Not Attempt Resuscitation (DNAR) decisions who are scheduled for surgery and anaesthesia. In the guidelines we used a separate section to define the rationale behind, and the implications of, a DNAR decision. The statement that ‘The term DNAR is intended to provide unambiguous instruction on one specific point – that a person should not undergo CPR in the event of cardiorespiratory arrest’ does not concur with our definitions. There are two problems with this statement. Firstly, although it is reasonable to assume that cardiopulmonary resuscitation will only ever be used following ‘spontaneous’ cardiopulmonary arrest in the out of hospital situation in which many DNAR decisions are implemented, there are in-hospital situations, including induction of anaesthesia, where cardiopulmonary resuscitation may be used when cardiopulmonary arrest has been induced by medical interventions. Cessation of spontaneous respiration, or apnoea, which frequently occurs following induction of anaesthesia, constitutes pulmonary arrest although this is not how we normally refer to it. The fact that it is iatrogenic and readily reversible, rather than a spontaneous event in the culmination of the dying process in a patient who has a terminal illness or poor quality of life, completely alters the assumptions under which the original DNAR decision was implemented. The cause of the cardiorespiratory arrest is highly relevant, and an ‘unambiguous instruction’ cannot be applied should it occur under anaesthesia. The second problem with the authors’ statement is that they argue that the DNAR decision to withhold cardiopulmonary resuscitation only becomes relevant after cardiopulmonary arrest has occurred. If this was the case, then it implies that the administration of cardiopulmonary resuscitation before cardiopulmonary arrest would not contravene a patient’s DNAR decision. This is erroneous and represents a misunderstanding of the rationale and implications of a DNAR decision. We do not believe this viewpoint should be adopted in the UK, nor should it be adopted to justify the administration of many aspects of anaesthesia, ‘which happen also to be components of CPR’ as the authors correctly state, to a patient undergoing surgery with an unmodified pre-existing DNAR decision. The purpose of publishing the guidelines was to assist doctors and patients by placing, in one document, the current ethical and legal principles involved in defining the best interests of the patient with a pre-existing DNAR decision who requires surgery, and to suggest how these may be applied to this rare but difficult situation. In doing so, we sought to protect patients from denial of their human rights as defined in the Human Rights Act 1998. We also sought to protect health professionals from retrospective accusations of assault arising from resuscitative interventions for which consent had not been granted, and from assisting euthanasia by pharmacologically inducing cardiopulmonary instability and then actively withholding cardiopulmonary resuscitation which, in other circumstances would have been indicated. To administer cardiopulmonary resuscitation to a patient who has explicitly refused such interventions in advance, even though the cardiopulmonary instability and/or arrest is pharmacologically induced, leaves an anaesthetist vulnerable to retrospective accusations of assault that may not be defensible and that may lead to criminal prosecution and referral to the General Medical Council (GMC). Under anaesthesia and surgery, the assumptions on which any pre-existing DNAR decision is made no long apply, and that is why all DNAR decisions should be reviewed in advance of surgery and anaesthesia. If the patient, proxy decision maker or doctor in charge of the patient decides that full cardiopulmonary resuscitation and its possible consequences during surgery and anaesthesia are in the patient’s best interests, then option 1 applies. If not, agreement must be made on what, if any, resuscitative interventions are acceptable, which is option 2. Agreeing to provide anaesthesia to a patient with an unmodified DNAR decision in place (option 3) would require cardiopulmonary resuscitation to be withheld, either as part of routine anaesthetic management or in the treatment of pharmacologically induced respiratory arrest. We believe that this is not in the patient’s best interests and certainly leaves the anaesthetist vulnerable to accusations of assisting euthanasia, which is currently illegal in the UK, and which may lead to criminal prosecution and GMC referral. Therefore, we believe that in most circumstances option 3 is incompatible with general anaesthesia for any type of surgery. Similar 3-option approaches have been adopted by the American Society of Anesthesiologists [1] and the Canadian Anesthesiologists’ Society [2]. We have attempted to provide guidelines that protect both patient and anaesthetist. We thank Smith and Poplett for their challenging letter, and hope this explains the rationale behind the 3-option approach recommended in the guidelines.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.