It is the final call after a long shift. Two paramedic crews respond to an elderly gentleman, with a history of coronary artery grafts and atrial fibrillation, found unresponsive and breathing abnormally by a neighbour who became alarmed when the door was not answered, despite the upstairs room light being on. On arrival of the crew, the patient had agonal respirations, the skin was mottled, and was pale. Chest compressions were commenced immediately. In the meantime, the kind neighbour rang a distraught family member, and a member of the crew spoke to them in an attempt to clarify the past medical history of the patient, as well as family and patient wishes.
Vomit was suctioned from the airway and ill-fitting dentures scooped out by the paramedics’ gloved finger. The defibrillator was attached and showed the patient to be in a non-shockable rhythm, with chest compressions only briefly stopping to allow this brief rhythm check.
One crew member prepared to cannulate a vein to deliver adrenaline, while another prepared to intubate. The family wanted everything possible done for their 91-year-old father. Chest compressions continued.
Cardiopulmonary resuscitation in elderly patients
Cardiopulmonary resuscitation (CPR) has the goal of returning patients with out-of-hospital cardiac arrest (OHCA) to independent living. Therefore, CPR in elderly patients requires careful consideration.
Globally, the population is ageing (World Health Organization (WHO), 2011). Projections based on a series of assumptions that take into account recent trends in fertility, mortality and migration, predict that one in four Australians will be 65 years or older by 2056, and those aged 85 years or over will make up 5% to 7% of Australia's population; compared with only 1.6% currently (Australian Bureau of Statistics, 2006-2011). Similar trends are predicted elsewhere.
Attitudes toward resuscitation in elders have been shown to vary, based on the perceived likelihood of a successful outcome (Russell et al, 1991; Carton and Brown, 1993; Schonwetter et al, 1993; Murphy et al, 1994; Hillier et al, 1995). What informs these perceptions for paramedics is not entirely clear; whether it is their level of experience, their knowledge, the influence of recent experiences with similar scenarios, their branch culture, the receiving hospital culture, or individual attitudes is unknown. However, it is important that the paramedic has available age specific results on what the likelihood is of their patient who they are deciding to resuscitate achieving return of spontaneous circulation (ROSC) and, more importantly, surviving to leave hospital. Armed with such information, it is much easier to talk with distraught families, and this background information may help justify why an attempted resuscitation is proceeding or discontinuing (Hakim et al, 1996; Phillips et al, 1996).
Generalizability of studies on out-of-hospital cardiac arrest in the elderly
Many investigators have studied outcomes after cardiac arrest in elders, with varying results (Ritter et al, 1985; Tresch et al, 1988; Murphy et al, 1989; Tresch et al, 1989; Van et al, 1992; Bonnin et al, 1993; Joslyn et al, 1993; Wuerz and Holliman, 1993; Tresch et al, 1994; Wuerz et al, 1995; Kim et al, 2000; Swor et al, 2000). Some have suggested that resuscitation for elders with cardiac arrest is rarely effective (Murphy et al, 1989; Podrid, 1989; Hazzard, 1989); it has even been described as a ‘curse’ of the elderly (Podrid, 1989). The limitations of these studies deserve consideration.
Certain studies did not take into account the Utstein (Langhelle et al, 2005) variables that help predict outcome, while other studies combined in-hospital and OHCA—this is akin to comparing apples with oranges as the patient groups may be completely different. Many previous studies have small sample sizes, and use nonstandard definitions of outcome; some report just ROSC, however survival to hospital discharge is more meaningful.
The age definition of ‘elderly’ has varied in the literature also. Importantly, most studies are based on data from cardiac arrests occurring in the 1990s, yet cardiac arrest outcomes have improved since, and much has developed in resuscitation science and how cardiopulmonary resuscitation is delivered. These limitations make meaningful comparisons difficult.
What are the outcomes for out-of-hospital cardiac arrest in elderly age groups?
We recently performed a study examining this question in Melbourne, Australia (Deasy et al, 2011a). Although it was retrospective and may have been influenced and biased by clinicians’ perception of likelihood of survival of the elderly patient in OHCA, the results help give an indication of OHCA outcomes in the elderly.
Appreciating the context and the standard of care provided when interpreting results from any study is important before deciding whether they are generalizable to another population. Australia has the fourth highest life expectancy for men and women in the world (WHO, 2005; Australian Institute of Health and Welfare, 2006). Ambulance Victoria is the sole provider of hospital care in Melbourne. Emergency call takers use the Advanced Medical Priority Dispatch System© and Ambulance Victoria are the sole provider of ambulance services. It is a paramedic provided service, with paramedics having a university degree in paramedicine.
Ambulance paramedics have some advanced life support skills—they are trained to place laryngeal mask airways and administer epinephrine, and mobile intensive care ambulance (MICA) paramedics are authorized to perform endotracheal intubation and administer a wider range of cardiac drugs. MICA paramedics are dispatched to patients with critical illness, including patients with cardiac arrest. In addition, a firefighter first responder programme operates for cardiac arrests in Melbourne's CBD area. The prehospital cardiac arrest protocols follow the recommendations of the Australian Resuscitation Council.
Selected data from the Victorian Ambulance computer information system (VACIS), an electronic patient record, and previous to this, the paper based patient care record, is collected from patients in cardiac arrest and stored on the Victorian Ambulance cardiac arrest registry (VACAR). The VACAR also includes some data from the hospital record for those patients transferred to hospital, including outcome. Specific circumstances exist under which Ambulance Victoria paramedics are permitted not to attempt resuscitation or may cease resuscitation. Reasons for not resuscitating may include presence of rigor mortis and obvious signs of death, injuries incompatible with life, or death declared by a medical officer at the scene.
Paramedics may discontinue efforts if, after 30 minutes of advanced life support, there has been no ROSC, the patient is not in ventricular fibrillation (VF) or ventricular tachycardia (VT), and has no signs of life present such as gasps or pupil reaction, and hypothermia or drug overdose are not suspected. The fourth link in the chain of survival (Nolan et al, 2006)—the post-resuscitation link which includes the decision to proceed with admission to intensive care, use of therapeutic hypothermia, and cardiac catheterization—clearly influences survival to hospital discharge rates.
There has been an appreciation of post-resuscitative care in Melbourne—one of the seminal trials in therapeutic hypothermia (Bernard et al, 2002) was performed here. However, the exact extent to which age influenced the emergency physician's, intensivist's or cardiologist's aggressiveness towards post-resuscitative care during the period that this study represents is difficult to ascertain.
Our study, using data derived from the VACAR, found that between 2000 and 2009, paramedics in Melbourne attended 30 006 out-of-hospital cardiac arrests; of which, 9609 were in the 65–79 year old age group, 6430 were octogenarians, 1530 were nonagenarians, and 40 were centenarians. Paramedics attempted resuscitation in 48% of 65–79 year olds, 39% of octogenarians, 31% of nonagenarians, and 17% of centenarians.
Outcomes for octogenarians and nonagenarians who sustain out-of-hospital cardiac arrest
In Melbourne, where OHCA resuscitation was performed by paramedics, ROSC and transfer to hospital occurred in 36% of 65–79 year olds, 31% of octogenarians and 20% of nonagenarians. However, ROSC does not tell the full story as many patients die after arrival to the hospital. There were 8% of 65–79 year olds, 4% of octogenarians, and 2% of nonagenarians who survived to leave hospital.
It was found that, although octogenarians were less likely to survive than their 65–79 year old counterparts, they had a 10.5% survival rate when the initial rhythm was shockable. Of nonagenarians found in a shockable rhythm where resuscitation was attempted, 4% survived to leave hospital alive. There was one centenarian found in ventricular fibrillation who represents the one survivor in the centenarian age group having been successfully defibrillated.
Over the last decade in patients aged 65 years or older, ROSC rates have improved for shockable and non-shockable rhythm OHCAs, although survival to hospital discharge has improved in the shockable rhythm group only (Deasy et al, 2011).
Survival to leave hospital alive is important, but quality of survival must be considered. VACAR has commenced telephone follow-up of all OHCAs, however, this data was not available for the time period of this study.
Out-of-hospital cardiac arrest in residential aged care facilities
The most common place for paramedics in Melbourne to encounter an elderly person in cardiac arrest was in their own home. In the 65–79 year old age group, 82% of OHCAs occurred at home; for octogenarians, the proportion was 78%, for nonagenarians 67%. However, an increasing number of OHCAs occur in residential aged care facilities, as age increases. In Melbourne, 6% of OHCAs in the 65–79 year old age group occurred in a residential aged care facility (RACF), while 16% of OHCAs in octogenarians and 29% of nonagenarians OHCAs occurred in such facilities (Deasy et al, 2011). Given that patients who enter residential aged care facilities usually do so due to increased infirmity and frailty, it is not surprising that paramedics are less likely to attempt resuscitation in those over 65 years who sustain their OHCA in RACF.
We recently analyzed OHCAs occurring in RACF in Melbourne (Deasy et al, 2011). The rate of bystander CPR administered in RACFs in our study was 32%, seemingly low for a location where health professionals would be expected to perform CPR. This may be due to a number of factors; RACF staff may have thought it inappropriate to perform chest compressions and were calling emergency services simply for support in their decision, perhaps it reflects a lack of awareness on behalf of RACF staff as to what it is an EMS is mandated to do in such an event, or it may identify a lack of BLS training and/or education in the area of death and dying for RACF staff.
We found paramedics 16% less likely to attempt resuscitation if the location of the OHCA was a nursing home, than in the patient's home. Differences in survival between OHCA occurring in the home and those occurring in a nursing home were noteworthy—survival to hospital discharge for non-shockable rhythm cardiac arrests occurring in nursing homes was 74% worse than when the arrest occurred in the patient's home.
However, while outcomes from OHCA occurring in residential aged care facilities are poorer than those occurring elsewhere, resuscitation is not futile with survival to leave hospital rates of 8.7% for shockable rhythm, and 1.2% for non-shockable rhythm where paramedics attempted resuscitaition.
Informing patients, their families and caregivers on issues around end-of-life and allowing an informed and appropriate decision be made regarding whether they would want an attempted resuscitation is a challenge for nursing homes, and the general practitioners and all health professionals that serve them.
Ideally, the decision to withhold CPR should be an informed consensus plan made by the patient in collaboration with his/her physician and family members with consideration of the underlying condition and the chances of survival.
Factors affecting the decision to resuscitate
Outcomes of OHCA are influenced greatly by the paramedic's decision to attempt or not attempt resuscitation. Most emergency medical services work to strict protocols, but a certain degree of discretion and subjectivity may be seen in the area of cardiac arrest resuscitation in the elderly patient who has presented in a non-shockable rhythm. Some paramedics may decide to initiate resuscitation in most cases, which will result in a lower rate of success as the denominator is increased. A high success rate may be explained by a restrictive attitude towards resuscitation.
The chances of successful survival after cardiac arrest depend upon co-morbidity, education, and organization of the health care services and emergency medical system. Known prognostic factors are the initial ECG-rhythm, time from collapse to first defibrillation, witnessed arrest, and bystander CPR. The economic resources available and demographic compositions of the population are also important indicators (Horsted et al, 2004).
We were interested in identifying what the patient and arrest characteristics were that motivated a paramedic decision to perform an attempted resuscitation in the older age groups, where the presenting rhythm was non-shockable. We found the factors associated with the decision to proceed with an attempted resuscitation in non-shockable OHCA were the presence of bystander CPR on arrival of EMS, a witnessed collapse, shorter EMS response time and OHCA occurring in a public place and younger age.
When we analyzed those non-shockable OHCAs occurring in residential aged care facilities, we found likewise—the adjusted odds of paramedics attempting resuscitation increased with the OHCA being witnessed and the patient receiving bystander CPR (Deasy et al, 2011). We did not find evidence of a change in the adjusted likelihood of a paramedic attempting resuscitation in the elderly over the decade.
As the number of elderly persons increase, the dilemmas of who to resuscitate and when increase. The decision can be highly emotional, stressful and complex, made even more difficult by the lack of reliable outcome data and in particular quality of life outcome data. In the ideal world, a well informed sensible plan would be made by the patient with the GP, in collaboration with family members well in advance of the event.
However, this is rarely the reality with EMS and emergency department personnel who are often confronting difficult decisions to initiate or withhold resuscitation. The American Heart Association (AHA) has long maintained that ‘except in narrowly defined circumstances… professional first responders are expected to always attempt BLS (basic life support) and ACLS’ (Cummins, 1994). The AHA's most recent ethical guidelines reiterate this imperative: you must start CPR (American Heart Association, 2005). The imperative to always resuscitate offers clarity but can have substantial emotional, moral, and financial costs.
EMS performance indicators
There is an opinion that because of the greater risk of harm and the lesser potential for benefit in older patients and the greater variability in treatment effects, the quality of evidence supporting substantial benefit and limited harm ideally should be more stringent in older vs younger populations (Scott and Guyatt, 2010).
Most services understandably measure success of treatment of the elderly in OHCA based on ROSC rates and survival; measuring the quality of death, the dignity offered to the dying elderly patient and the quality of acute bereavement support for their families has not been a focus. Likewise, to date, there has been a lack of robust reporting of functional and quality of life outcomes; these are integral to painting the true picture of EMS performance.
‘Outcomes of OHCA are influenced greatly by the paramedic's decision to attempt or not attempt resuscitation’
Conclusions
The definition of medical futility has been considered as a survival rate of less than 1% (Schneiderman et al, 1990), although this has been debated (Ardagh, 2000; Helft et al, 2000). OHCA outcomes, even in the very elderly, show it is not futile.
However, there is a real need for a community-wide approach to expected natural deaths occurring outside of hospital. This would result in improvements in end-of-life care, preserving patient autonomy, diminishing suffering, and avoiding unwanted, inappropriate resuscitative efforts, if carried out appropriately within a supportive EMS infrastructure.