End-of-life in the ICU: moving from 'withdrawal of care' to a palliative care, patient-centred approach.

End of life decisions in the Intensive Care Unit [ICU] are difficult for patients, families and doctors alike, yet they are increasingly common; ICNARC data (UK) shows that 15–25%of those admitted will die in the ICU and European data shows approximately 70% of these deaths occur after the withholding or withdrawal of lifesustaining treatments. Between 10 and 20% of the population at large now die in ICU underlining the importance of end-of-life care to everyday practice and training for anaesthetists and intensivists. Despite how common end-of-life decisions are in ICU, they are very variable, with studies showing important differences reported between regions, countries, individual ICUs and even between individual clinicians practicing in the same ICU. 3 This was demonstrated by a recent systematic review; its accompanying editorial introduced the concept that these practice variations may be good or bad. Good variations reflecting patient-centered care, and bad variation reflecting failures in professionalism. Currently the ETHICUS 2 (world-wide) study is underway, collecting prospective, data on ICU end-of-life practices, in a manner similar and comparable with the initial ETHICUS study, which occurred in 1999–2000. Consequently, the results of ETHICUS 2, will likely reflect the practice changes of the past 16 yr and the resultant changes in both good and bad variations. For example, high profile end-of-life cases in Europe have generated intense media debate leading to increased public awareness of withholding and withdrawal of life-sustaining treatments, and the Liverpool Care Pathway has caused a re-examination of aspects of hospital end-of-life practices and emphasized the importance of appropriately integrated palliative care strategies. Terminology confusion however, has been slowing progress in qualityend-of-life care. Forexample, three Europeanstudies examined end-of-life practices in ICU in the period 1995–2000 and each of them had differences in their definitions of thewithholding and withdrawal of life-sustaining treatments. 7 8 TheWELPICUS study however, has achievedworld-wide consensus on keyend-of-life issues and terminology. Using a Delphi technique requiring 80% agreement, 35 definitions and 46 consensus statements regarding 22 end-of-life ICU issues were processed. Agreement was reached on the majority of these definitions and statements which included the ‘withholding and withdrawal of life-sustaining treatments’ (see Table 1). The WELPICUS consensus now provides health-care professionalswith terminology for everyday purposes, thereby limiting previous confusing variations. However, whether ‘withholding’ and ‘withdrawal’ are ethically equivalent is debated. Equivalence implies that if a treatment, (e.g. mechanical ventilation) is disproportionately burdensome for the patient, in that it will offer no clinical improvement and/or may prolong suffering, then regardless of whether ventilation that is already ongoing is stopped (withdrawal), or not started from the outset (withheld), the principle (preventing prolonged patient suffering via a non-beneficial therapy) is regarded as the same. This is supported in the guidelines of most critical care societies and medical regulatory bodies.