Evolution of End-of-Life Practices in a Spanish Intensive Care Unit between 2002 and 2009

Abstract

Background:

Spanish cultural perception of end-of-life decision making has started to change within a new legal framework.

Objectives:

The objective of this study was to describe how life-sustaining treatments (LST) were withheld or withdrawn in an intensive care unit (ICU), to determine the degree of agreement between physicians and patients' families on end-of-life decisions, and to compare our results to those of studies undergone before these legal changes.

Design:

This was an observational retrospective study.

Setting/Subjects:

The setting and subjects were patients admitted to a medical and surgical ICU in a metropolitan tertiary care center from January 2002 to December 2009 whose LST had been withheld or withdrawn.

Measurements:

Study measurements included age and sex; comorbidities; functional status on admission; main diagnosis; SAPS II and APACHE II scores on admission; degree of agreement on end-of-life practices; and therapeutic interventions withheld or withdrawn and outcome.

Results:

A total of 371 of 6526 patients (5.7%) had LST withheld or withdrawn; 339 of these patients died in the ICU. Patients whose LST was withheld or withdrawn were older, had a high number of comorbidities, and were admitted with higher SAPS II and APACHE II scores than the general ICU population. Active treatments and basic support were discontinued in 212 patients (57%) and 100 patients (26.9%), respectively. An agreement between the staff and the patient's family was reached in 318 cases (85.7%). Families were not involved in 30 cases (8.1%).

Conclusions:

Compared to prior studies, shared end-of-life decision making in the ICU has increased in recent years. Decisions to forgo treatments mainly involve advanced life support.

Introduction

Medical practices for end-of-life care in the intensive care setting vary greatly around the world.¹ The reasons for this variability are diverse and include differences in geographic setting, cultural and religious values, legislation, and personal attitudes of physicians.^2,3 A nationwide survey carried out in the United States showed that 70% of deaths in the ICU setting were preceded by a decision to forgo treatment.⁴ Close to 80% of ICU deaths in a Canadian cohort had life-sustaining therapies (LST) withheld or withdrawn.⁵ Most observational studies from European countries also concur on the high variability of these practices, and highlight the absence of a consensual approach to this area of medical care.^1,6,7 International critical care societies advocate shared decision making about withholding or withdrawing LST in the critical care setting.⁸ Substantial geographic differences also exist in the degree of involvement of the family in the end-of-life decision-making process. Observational studies at the beginning of the past decade in Spain and France showed that 44% to 72% of families were involved in this process,^6,7 whereas they were more commonly involved in northern European or Asian countries.^2,9

In the last decade, Spanish cultural perception of end-of-life decision making has started to change. The enactment of a new law regulating patient autonomy and rights in 2002 reflected the triumph of a novel autonomy-driven doctor-patient relationship over the prior paternalistic model.¹⁰ In essence, the law regulates the need and uses of informed consent and advance care directives, and the protection of the information contained in the patient's medical record. In the following years, local critical care scientific societies issued specific recommendations for end-of-life care in the ICU.^11–13 The impact of this legal setting on end-of-life decisions within the Spanish health care system has only been marginally assessed.^14–17 Our primary objective is to describe the frequency with which LST was withheld or withdrawn between 2002 and 2009. Secondary objectives are to determine the degree of agreement between physicians and patients' families (or surrogate decision makers) on end-of-life decisions during that same period, and to compare these results to prior studies, undergone before these legal changes took place.^7,18

Methods

This is a retrospective observational study of patients admitted to the ICU of a single tertiary care center in Madrid, Spain from January 2002 to December 2009, in whom some treatment measures (including but not limited to LST) had been withheld or withdrawn during their admission. The 20-bed ICU cares for medical and surgical critically ill patients. Since the beginning of 2002, all ICU admissions were recorded in a computerized database that enables patients' electronic records to be identified by their code status at their ICU discharge (full support, do-not-resuscitate, or withholding/withdrawing treatment). Patients who fulfilled brain death criteria were excluded from the study.^19,20 The local institutional review board approved the study protocol and waived the need for informed consent owing to the lack of intervention and the retrospective nature of the study. All data were treated with the highest confidentiality, according to the Spanish data protection act.²¹

Definitions

Treatments were classified according to their different impact on the overall clinical status of the patient, and separated into three mutually exclusive categories.

(1) Advanced life support (also referred to as LST) was defined as any medical or technical interventions, procedures, or medications administered to a patient with the primary intention of delaying death, regardless of whether these were also administered to treat the underlying illness.²² Advanced life support included mechanical ventilation, high-content oxygen therapy (FiO₂ higher than 60%), vasoactive drugs (norepinephrine, epinephrine, dobutamine, or dopamine); dialysis; and/or transfusions (of any blood product).

(2) Active treatment was defined as all interventions, procedures, or medications administered to a patient with the primary intention of treating the underlying condition and included antibiotics, surgery, or insulin, among others.

(3) Basic support included hydration (more than 1000 cc of intravenous fluids per day or equivalent oral amount) and nutrition (enteral or parenteral).

Withholding treatment was defined as the decision not to institute a medical therapy, either based on the patient's autonomous refusal of treatment, or considering that the treatment, though medically appropriate and potentially beneficial in habitual cases, would be unable to modify the patient's outcome. Do-not-resuscitate orders were considered in this category.

Withdrawing treatment was defined as the cessation and removal of an ongoing medical therapy with the explicit intention not to provide an equivalent alternative treatment, knowing that the patient may die as a result of the change in approach. Patients who had both withholding and withdrawal of treatment measures were classified as having withdrawal.^1,4,7 Only those end-of-life decisions whose rationale was explicitly recorded in the medical records were included in the analysis. Comfort care measures were provided in all cases after an end-of-life decision was met.

Regarding the degree of agreement between the physician and the patient's family, all records were reviewed for explicit comments on the results of any meeting concerning end-of-life decisions (either “family–staff” or “staff only” meetings). All proposals to withhold or withdraw treatment were discussed within the medical staff. For the analysis, the following mutually exclusive categories were used:⁷

(1) Direct agreement: The physician and the patient's family discussed the patient's prognosis and agreed that medical therapy was unlikely to restore the patient to health. They then decided either not to administer or to remove medical interventions with the understanding that the patient's death might occur as a result. This agreement could also be reached concerning just one arm of the decision-making process (withholding or withdrawing treatment).

(2) Indirect agreement: Prognostic information was provided, and the physician and patient's family agreed on the futility of medical therapy, although decisions to withhold or withdraw specific interventions were not overtly expressed. The medical staff decided afterwards which measures were susceptible for being withheld or withdrawn.

(3) Family proposal: The decision to forgo medical measures was first proposed by the patient or family. This category included the use of advance care directives.

(4) Family rejection: The patient or the patient's family rejected the initial proposal from the medical staff to forgo life support.

(5) No consultation: The decision to withhold or withdraw life support was taken exclusively by the ICU medical staff.

Measurements

All medical records were reviewed by the same investigator (DRA). We collected demographic variables (age and gender); clinical variables (date of admission, original ward [medical, including the emergency department, or surgical]; prior comorbidity, assessed using the age-adjusted Charlson index;^23,24 the main ICU diagnosis and clinical situation on admission based on the Simplified Acute Physiology Score II [SAPS II] and the Acute Physiology and Chronic Health Evaluation II scale [APACHE II]^25,26); and end-of-life care variables (date of the decision, degree of agreement on end-of-life practices, patient's clinical situation at the time of the decision [SAPS II and APACHE II], treatments withheld or withdrawn, evolution after the decision, and date of discharge).

Statistical analysis

All data were processed using SPSS 20.0 (IBM Corp., Armonk, NY). A χ² test (with Yates correction when applicable) was used to evaluate the statistical significance of categorical variables, which are presented as absolute frequencies and percentages. Quantitative variables are presented as mean±standard deviation (SD), and were compared with Student's t test or the Mann-Whitney U test. Time-dependent variables were expressed as median and interquartile range (IQR). Trends over time were analyzed with the nonparametric Mann-Kendall test. All statistical tests were two-tailed. Results with p<0.05 were considered significant.

Results

A total of 6526 patients were admitted to the ICU from January 2002 to December 2009 (816±45 admissions per year). Of these, 1029 (16.7%) died in the ICU. A decision to forgo treatment was made in 371 patients (5.7% of all admissions), of whom 339 died (91.4% of the study sample, 32.9% of all ICU deaths). Baseline characteristics of the patients in whom an end-of-life decision was reached are shown in Table 1. Compared with the general ICU population, these patients were older, came more frequently from medical wards, had a high number of medical comorbidities, and were admitted to the ICU with a poorer clinical status (p<0.01 for all comparisons). Their median length of stay in the ICU was also significantly longer: eight days (IQR 3–21 days) versus three days (IQR 1–8 days); p<0.01. The main diagnostic categories at admission are listed in Table 2. The most common reasons for ICU admission were cardiovascular disease (including patients with cardiac arrest) and neurological disease, both in the general ICU population and in the group whose treatment was subsequently withheld or withdrawn.

Table 1.

Baseline Characteristics of Patients Whose Treatment Was Withheld or Withdrawn during Their ICU Admission

Clinical characteristics	General population (n=6526)	WH/WD (n=371)	p
Age (years)	59.5±16.4	68.1±13.4	<0.01
Gender (% males)	4063 (62.2%)	239 (64.4%)	NS
Original ward (% medical patients)	2630 (40.3%)	299 (80.6%)	<0.01
Charlson index	N/A	4.8±3.6	N/A
SAPS II at admission	36±18	56±19	<0.01
APACHE II at admission	17±8	26±8	<0.01
Length of stay in ICU (days)	3 (1–8)	8 (3–21)	<0.01

Results are presented as mean±SD or n (%).

APACHE II, Acute Physiology and Chronic Health Evaluation II; N/A, data not available; NS, not significant; SAPS II, Symplified Acute Physiology Score II; WH/WD, patients whose treatment was withheld or withdrawn.

Table 2.

Main Admitting Diagnosis of Patients in Whom an End-of-Life Decision Was Made

Diagnostic category	General population (n=6526)	WH/WD (n=371)	p
Cardiovascular	3371 (51.7%)	160 (43.1%)	<0.01
Post-CPR		106 (28.6%)
Cardiac or major vascular surgery		42 (11.3%)
Pulmonary edema and other cardiological diagnosis		12 (3.2%)
Neurological	1193 (18.3%)	62 (16.7%)	NS
Medical^a		57 (15.4%)
Surgical^b		5 (1.3%)
Respiratory	607 (9.3%)	42 (11.3%)	NS
Infectious diseases	465 (7.1%)	27 (7.3%)	NS
Gastrointestinal	366 (5.6%)	26 (7.0%)	NS
Other^c	524 (8.0%)	54 (14.6%)	<0.01
Hemato-oncological		34 (9.1%)

Results are presented as n (%).

Includes ischemic or hemorrhagic strokes, seizures, and nonsurgical cranial trauma.

Includes surgical cranial hemorrhages (traumatic or not) and brain tumors.

Includes patients with severe trauma (cranial trauma excluded), renal, urological, metabolic, toxic, and/or oncological illnesses.

CPR, cardiopulmonary resuscitation; NS, not significant; WH/WD, patients whose treatment was withheld or withdrawn.

Evolution of end-of-life practices

During the study period, 371 patients had some treatment withheld or withdrawn. The annual rate of patients whose treatments were forgone, as well as their outcome, is summarized in Table 3. We found no significant interannual differences in the proportion of patients whose treatments were withheld or withdrawn throughout the study period. This decision was mostly reached during the first week of admission, with a median time of four days (IQR 1–13 days). SAPS II and APACHE II scores at the time of decision were significantly lower than those at admission to the ICU—SAPS II: 53±18 points versus 56±19 points, 95% CI (-4.9 to−1.1); and APACHE II: 23±8 points versus 26±8 points, 95% CI (-3.8 to -2.2); p<0.01 for both comparisons. After the decision, death occurred most frequently in the following 96 hours (median time 1 day, IQR 0–4). A total of 302 patients (81.4%) died in the ICU, while 37 (10%) died in the hospital after discharge from the ICU. Thirty-two patients (8.6%) in whom a decision to withhold/withdraw treatment was made were discharged from the hospital after a median length of stay of 31 days (IQR 12–57 days).

Table 3.

Annual Distribution and Clinical Evolution of Patients Whose Treatment Was Withheld or Withdrawn during Their ICU Admission ^a

Patient category	2002	2003	2004	2005	2006	2007	2008	2009	p
All patients	829	718	850	841	853	819	833	783	NS
All deaths^a	157 (18.9%)	154 (21.4%)	146 (17.2%)	148 (17.6%)	141 (16.5%)	108 (13.2%)	118 (14.2%)	120 (15.3%)	NS
WH/WD^b	43 (5.2%)	58 (8.1%)	60 (7.1%)	51 (6.1%)	56 (6.6%)	40 (4.9%)	31 (3.7%)	32 (4.1%)	NS
WH/WD deaths^c	42 (26.7%)	55 (35.7%)	52 (35.6%)	46 (31.1%)	53 (37.6%)	36 (33.3%)	28 (23.7%)	27 (22.5%)	NS
WH/WD discharged from hospital^d	1 (2%)	3 (5%)	8 (13%)	5 (10%)	3 (5%)	4 (10%)	3 (10%)	5 (16%)	NS

Results are presented as n (%).

NS, not significant; WH/WD, patients whose treatment was withheld or withdrawn.

Overall, 354 patients (95.4%) had advanced life support withheld or withdrawn. Advanced life support was maintained in eight patients (2.2%), despite a decision to withhold active treatment. In these eight cases the decision involved not performing life-saving surgery, based on the patient's clinical status. Active treatments were also frequently discontinued (212 patients, 57%); but basic support measures were mostly maintained (262 patients, 71%). Table 4 summarizes the decisions to withhold/withdraw treatment based on treatment categories. The most frequently withdrawn interventions were vasoactive drugs (149 patients, 40%); antibiotics (115 patients, 31%); and other active treatments (e.g., surgery was withheld in 127 patients, 34%). Mechanical ventilation was withdrawn in 107 cases (29%). As described in Table 5, we found no statistically significant differences in the frequencies with which each advanced life support measure was withheld or withdrawn. No significant interannual trends in how treatments were withheld or withdrawn were observed.

Table 4.

Distribution of the Decision to Withhold or Withdraw Treatment in 371 ICU Patients According to Treatment Categories ^a

Treatment category	Maintained	WH/WD	N/A
Life-sustaining therapies	8 (2.2%)	354 (95.4%)	9 (2.4%)
Active treatment	147 (39.6%)	212 (57.1%)	12 (3.3%)
Basic support	262 (70.6%)	100 (26.9%)	9 (2.5%)

Results are presented as n (%).

N/A, not available from the clinical record; WH/WD, treatment withheld or withdrawn.

Table 5.

Distribution of the Therapeutic Interventions Withheld or Withdrawn in 371 Critically Ill Patients ^a

Treatments	Maintained	Withheld	Withdrawn	N/A
Mechanical ventilation	126 (34%)	116 (31.3%)	107 (28.8%)	22 (5.9%)
High FiO₂	98 (26.4%)	101 (27.2%)	96 (25.9%)	76 (20.5%)
Vasoactive drugs	50 (13.5%)	67 (18.1%)	149 (40.1%)	105 (28.3%)
Dialysis	6 (1.6%)	77 (20.7%)	39 (10.5%)	249 (67.1%)
Transfusions	24 (6.5%)	52 (14%)	32 (8.6%)	263 (70.1%)
Antibiotics	143 (38.5%)	31 (8.4%)	115 (31%)	82 (22.1%)
Other active treatments	137 (36.9%)	39 (10.5%)	127 (34.2%)	68 (18.3%)
Nutrition	213 (57.4%)	17 (4.6%)	74 (19.9%)	67 (18.1%)
Hydration	231 (62.3%)	8 (2.2%)	64 (17.2%)	68 (18.3%)

Results are presented as n (%).

N/A, not available from the clinical record.

Degree of agreement between physicians and families over end-of-life practices

End-of-life decisions were explicitly recorded for 356 of 371 patients (95.9%). The proposal to withhold or withdraw life support was initiated by physicians in 336 patients (94.4%). No patients had decision-making capacity at the time. In 243 patients (65.5%), both the family and the physicians decided to withhold or withdraw life support (direct agreement). An indirect agreement was reached in 55 cases (14.8%). The decision was based on the patient's advance directives or on a family proposal in 20 cases (5.4%). The family was not involved in the decision to withhold or withdraw treatment in 30 patients (8.1%). The rationale for these unilateral decisions was not explicitly recorded in the medical charts. The family rejected the physician's first proposal to withhold or withdraw treatment in 8 patients (2.2%). In 6 of these cases, a subsequent agreement was reached in the following days. The other 2 patients died despite advanced life support in the following 24 hours. We found no significant interannual differences in the degree of physician-family agreement throughout the study period.

Discussion

One-third of ICU deaths in our critical care setting follow a nontreatment decision. Decisions to forgo treatment are made primarily over advanced life support. Compared to the general ICU population, these patients were usually older, came more frequently from medical wards, and had a high number of comorbidities before admission to the ICU. These factors have already been identified as predictors for higher mortality and are associated with a higher rate of withdrawal from mechanical ventilation.^27,28

The number of ICU patients whose treatment was withheld or withdrawn in our study is somewhat lower than those recorded in other European studies.^1,6,7 Although 73% of ICU deaths in the ETHICUS study were preceded by some limitation of treatment, the figure was substantially lower in southern European ICUs, where 40% of ICU deaths were preceded by a decision to withhold LST, and only in 18% was LST withdrawn.¹ A decision to forgo treatment was implemented in 50.4% of ICU deaths in the study by Pochard and colleagues.⁶ Although our results also differ from those observed in pediatric ICUs or in the general population,^29,30 they concide with those of Esteban and colleagues.⁷ This difference between Spanish and European end-of-life decisions' prevalences could have several reasons: advanced directives are still only marginally implemented in our sociocultural setting. Decisions to forgo LST are mainly reached after a proposal from the medical staff, and mostly after substantial evidence on the expected poor long-term prognosis of the patient is gathered. During this delay some patients who have undergone CPR or suffered a major neurologic insult could die despite LST, and before a complete discussion with the families could take place. The greater time from ICU admission until the first decision to limit treatment in our study compared to the ETHICUS cohort—4 days (IQR 1–13 days) versus 2.8 days (IQR 0.6–9.8 days)—is yet more indirect evidence to support this hypothesis.

Compared to earlier studies, we observed a substantial increase in the withdrawal or withholding of advanced life support, active treatments, and even basic support measures, perhaps indicating a more thorough and dynamic approach to the adjustment of treatment in critically ill patients during the last years. For instance, mechanical ventilation was withdrawn in 107 patients (29%) in our study compared to 19.5% in the study by Cook and colleagues,²⁷ 13.5% in the series by Diringer and colleagues,²⁸ or 9% in the ETHICUS cohort.¹ In a surprisingly high proportion of cases, active treatments and basic life support were continued in the context of ending advanced life support (39.6% and 70.6%, respectively). Forgoing active treatment or basic life support could have been considered unnecessary in some cases, when withholding/withdrawing advanced life support was expected to lead to death in the following hours—before the next dose of any planned active therapy, such as antibiotics, could be administered. The fact that basic support was only ended in 26.9% of cases could reflect the distinction between “nutrition and hydration” and other treatment categories that many health care professionals and families still make on moral or religious grounds. The difficulty in understanding the moral parity between withholding and withdrawing treatment may underlie many of these situations.^31–33

We observed a considerable increase in the rates of agreement between physicians and families regarding end-of-life decisions, when compared to prior studies in southern European countries. In this study, an agreement between physicians and family members was achieved in 318/371 cases (85.7%) and was mostly reached during the first four days after the admission. In a previous European survey, only 49% of respondents would involve medical staff, patients, and families in end-of-life decisions.³⁴ In an Italian survey, 19% of physicians said the close family was never involved in such decisions, and 56% would never involve patients even if competent.³⁵ Pochard and colleagues noted that health care proxies only participated in 17% of the cases.³⁶ Two Spanish studies, conducted before the enactment of the new patients autonomy law in 2002, also showed a lower rate of consensus (71.7% to 73.5%).^7,18 Other recent studies have also shown markedly lower rates of involvement of relatives.³⁷ These results could reflect both an improvement in professionals' communication skills, and an increased awareness of a new autonomy-based doctor-patient relationship in the last decade, according to which families now demand more active participation in end-of-life decisions.

The present study has several strengths and some potential limitations. Unlike earlier surveys and questionnaire-based studies, which may not illustrate daily clinical practice, our study was based on the review of medical charts.^4,6,34 To our knowledge, we provide the longest and largest single-center series to date on end-of-life decisions in the critical care setting. The same investigator reviewed all the charts and records and applied uniform evaluation criteria, thus making our results homogeneous, although a second reviewer could have also improved the reliability of the data extraction. Thirdly, the decision to withhold/withdraw treatment was recorded in an extraordinarily high number of cases.³⁸ The fact that only those end-of-life decisions whose rationale was explicitly recorded in the charts were included in the study may have partly accounted for the relatively low prevalence of end-of-life decisions in our setting. It is a single-center study, which may limit the generalization of our results. Physicians may have classified “direct agreement” cases more readily, thus generating an information bias, unfortunately inherent to all retrospective studies. Moreover, the nonsignificant reduction in the number of recorded cases in the last years of the past decade may have been caused by a decline in reporting, and not by a real decrease in these end-of-life practices. We have no information on the patients whose admission to the ICU was withheld, although this may be yet another approach to forgoing LST.³⁹

Our results underline the increasing importance of end-of-life decisions in the critical care setting. Decisions to forgo treatment are made primarily over advanced life support, although the futility of active treatments and basic support measures is also considered with increasing frequency. In the last years, these practices have involved patients and their families more actively. Although our present results are encouraging, more work must be done to foster broader deliberation on the need to promote full autonomous decisions at the end of a patient's life.

Footnotes

Acknowledgments

We would like to thank F. Rodríguez for his support with the statistical analysis and T. O'Boyle for his review of the final version of the manuscript.

Author Disclosure Statement

The authors have no conflicts of interest or specific funding sources to disclose.

References

Sprung

, Cohen

, Sjokvist

et al. End-of-life practices in European intensive care units. The ETHICUS study. JAMA, 2003; 290:790–797.

Curtis

, Vincent

. Ethics and end-of-life care for adults in the intensive care unit. Lancet, 2010; 375:1347–1353.

Sprung

, Carmel

, Sjokvist

et al. ETHICATT Study Group: Attitudes of European physicians, nurses, patients, and families regarding end-of-life decisions: The ETHICATT study. Intensive Care Med, 2007; 33:104–110.

Prendergast

, Claessens

, Luce

. A national survey of end-of-life care for critically ill patients. Am J Respir Crit Care Med, 1998; 158:1163–1167.

Hall

, Rocker

. End-of-life care in the ICU. Treatments provided when life support was or was not withdrawn. Chest, 2000; 118:1424–1430.

Ferrand

, Robert

, Ingrand

, Lemaire

. for the French LATAREA group: Withholding and withdrawal of life support in intensive-care units in France: A prospective survey. Lancet, 2001; 357:9–14.

Esteban

, Gordo

, Solsona

et al. Withdrawing and withholding life support in the intensive care unit: A Spanish prospective multi-center observational study. Intensive Care Med, 2001; 27:1744–1749.

Carlet

, Thijs

, Antonelli

et al. Challenges in end-of-life care in the ICU. Statement of the 5th international consensus conference in critical care, Brussels, Belgium, April 2003. Intensive Care Med, 2004; 30:770–784.

Mani

, Mandal

, Bal

et al. End-of-life decisions in an Indian intensive care unit. Intensive Care Med, 2009; 35:1713–1719.

10.

State Official Bulletin. Law 41/2002, of the 14th November, regulating the patient's autonomy and the rights and obligations concerning clinical information and documentation. www.boe.es/boe/dias/2002/11/15/pdfs/A40126-40132.pdf. 2012 November 3.

11.

Cabré

, Abizanda

, Baigorri

et al.

Code of ethics of the Spanish Society of Intensive Care, Critical and Coronary Units (SEMICYUC)

Med Intensiva, 2006; 30:68–73.

12.

Monzón

, Saralegui

, Abizanda

et al. Treatment recommendations at the end of the life of the critical patient. Med Intensiva, 2008; 32:121–133.

13.

Monzón

, Saralegui

, Molina

et al. Grupo de Bioética de la SEMICYUC: Ethics of the cardiopulmonary resuscitation decisions. Med Intensiva, 2010; 34:534–549.

14.

Tagarro

, Dorao

, Moraleda

et al. End-of-life care in a Spanish pediatric intensive care unit: Staff and parental evaluation. An Pediatr (Barc), 2008; 68:346–352.

15.

Esteve

, Jimenez

, Perez

et al. Factors related to withholding life-sustaining treatment in hospitalized patients. J Nutr Health Aging, 2009; 13:644–650.

16.

Ansuategui

. Euthanasia, philosophy, and the law: A jurist's view from Madrid. Camb Q Healthc Ethics, 2009; 18:262–269.

17.

Rubio

, Sanchez

, Fernandez

. Life-sustaining treatment limitation criteria upon admission to the intensive care unit: Results of a Spanish national multicenter survey. Med Intensiva, 2013; 37:333–338.

18.

Iribarren-Diarasarri

, Latorre-Garcia

, Muñoz-Martinez

et al. Limitation of therapeutic effort after ICU admission: Analysis of related factors. Med Intensiva, 2007; 31:68–72.

19.

Escalante

, Escudero

. Conclusion of the III Conference of Consensus of the SEMICYUC Brain death in critical care units. Med Intensiva, 2000; 24:193–197.

20.

Escudero

. Brain death diagnosis. Med Intensiva, 2009; 33:185–195.

21.

State Official Bulletin. Law 15/1999 of the 13th December, Personal Data Protection Act. www.boe.es/boe/dias/1999/12/14/pdfs/A43088-43099.pdf. 2012 November 3.

22.

Hastings Center. Guidelines on the Termination of Treatment and the Care of the Dying. Briarcliff Manor, NY: Hastings Center, 1987.

23.

Charlson

, Pompei

, Mackenzie

. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chron Dis, 1987; 40:373–383.

24.

Hall

, Ramachandran

, Narayan

et al. An electronic application for rapidly calculating Charlson's comorbidity score. BMC Cancer, 2004; 4:94.

25.

Le Gall

, Lemeshow

, Saulnier

. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA, 1993; 270:2957–2963.

26.

Knaus

, Draper

, Wagner

et al. APACHE II: A severity of disease classification system. Crit Care Med, 1985; 13:818–829.

27.

Cook

, Rocker

, Marshall

et al. Withdrawal of mechanical ventilation in anticipation of death in the intensive care unit. N Engl J Med, 2003; 349:1123–1132.

28.

Diringer

, Edwards

, Aiyagari

et al. Factors associated with withdrawal of mechanical ventilation in a neurology/neurosurgery intensive care unit. Crit Care Med, 2001; 29:1792–1797.

29.

Seale

. End-of-life decisions in the UK involving medical practitioners. Palliat Med, 2009; 23:198–204.

30.

Verhagen

, De Vos

, Dorscheidt

et al. Conflicts about end-of-life decisions in NICUs in the Netherlands. Pediatrics, 2009; 124:e112–e119.

31.

Callahan

. What Kind of Life. The Limits of Medical Progress. New York: Simon & Schuster, 1990.

32.

Cabré

, Solsona

. for the Spanish Society of Intensive Care Medicine (SEMICYUC) Bioethic study group: Limitation of therapeutic activity in intensive care medicine. Med Intensiva, 2002; 26:304–311.

33.

Beck

, Van der Loo

, Reiter-Theil

. A “little bit illegal”? Withholding and withdrawing of mechanical ventilation in the eyes of German intensive care physicians. Med Health Care Philos, 2008; 11:7–16.

34.

Vincent

. Forgoing life support in western European intensive care units: The results of an ethical questionnaire. Crit Care Med, 1999; 27:1626–1633.

35.

Giannini

, Pessina

, Tacchi

. End-of-life decisions in intensive care units: Attitudes of physicians in an Italian urban setting. Intensive Care Med, 2003; 29:1902–1910.

36.

Pochard

, Azoulay

, Chevret

et al. for the French PROTOCETIC group: French intensivists do not apply American recommendations regarding decisions to forgo life-sustaining therapy. Crit Care Med, 2001; 29:1887–1892.

37.

Jox

, Krebs

, Fegg

et al. Limiting life-sustaining treatment in German intensive care units: A multiprofessional survey. J Crit Care, 2010; 25:413–419.

38.

Sprung

, Woodcook

, Sjøkvist

et al. Reasons, considerations, difficulties and documentation of end-of-life decisions in European intensive care units. Intensive Care Med, 2008; 34:271–277.

39.

Walter

, Siegler

, Hall

. How decisions are made to admit patients to medical intensive care units (MICUs): Academic medical centers across the United States. Crit Care Med, 2008; 36:414–420.