Over the last two to three decades, the epidemiology and outcomes of out-of-hospital cardiac arrest (OHCA) have been studied in many countries (Westfal et al. 1996; Kuisma and Matta, 1996; Fischer et al. 1997; Waalewijn et al. 1998; Bottinger et al, 1999; Rewers et al, 2000; Finn et al, 2001; Eckstein et al. 2005; Estner et al, 2007; Fairbanks et al, 2007; Fridman et al, 2007; Grmec et al, 2007; Huei-Ming et al, 2007; Erdur et al, 2008; Shiraki et al, 2009; Yanawaga and Sakomoto, 2010). Mostly, these studies have focused on urban settings in developed countries which have well-resourced emergency medical service (EMS) systems. Little has been published about the nature of OHCA or associated outcomes in EMS systems of developing countries, where the resources available for EMS provision may differ markedly from those in the developed world.
Although survival from OHCA varies between locations, sometimes even within one country, (Nichol et al, 2008), characteristics of well-resourced EMS systems tend to be fairly consistent. In particular, these include efficient functioning of call centres that are able to rapidly dispatch EMS vehicles, and high availability of EMS vehicles in a specific geographic area.
Most published data on OHCA document short response times and early access to patients made possible by these two EMS characteristics (Westfal et al. 1996; Kuisma and Matta, 1996; Fischer et al. 1997; Waalewijn et al. 1998; Bottinger et al, 1999; Rewers et al, 2000; Finn et al, 2001; Eckstein et al. 2005; Estner et al, 2007; Fairbanks et al, 2007; Fridman et al, 2007; Grmec et al, 2007; Erdur et al, 2008; Shiraki et al, 2009; Yanawaga and Sakomoto, 2010). Although perhaps taken for granted in developed EMS systems, such characteristics may be lacking in less well-resourced systems.
A developing world example
South Africa is a developing country with an upper-middle-income and advanced emerging market economy (The World Bank, 2010; FTSE, 2010). In 2009, inflation-adjusted income per person in South Africa was US$ 9141. Despite this, life expectancy is low at 52 years (Gapminder, 2010).
South Africa has a high prevalence of injury, particularly due to inter-personal violence. The country's violent death rate is close to five times the worldwide average and injury overall is the second leading cause of death (Seedat et al, 2009). This, coupled with a large burden of infectious disease (the leading cause of death) including a high percentage of young adults with HIV, means that EMS throughout the country tend to be under-resourced for the large populations and high call volumes serviced, particularly in urban areas. (Wallis et al. 2008). Rapid urbanization and an ongoing influx of immigrants from neighbouring states means that population growth in urban centres is particularly rapid.
The largest proportion of this growth tends to be within the lowest socio-economic strata that are completely dependent on state-funded health care. Virtually all large cities in South Africa have burgeoning informal settlements where living conditions, sanitation and access to basic services are poor. Such conditions place a heavy burden on EMS systems whose resources seem unable to keep up with the growing demand for prehospital emergency care.
Prehospital emergency care in South Africa is provided by EMS's falling under the authority of provincial health departments in each of the nine provinces. Advanced life support level care has been a feature of South African prehospital care since the mid-1980s, with the current paramedic educational structure in the process of transition from a three-tiered system (basic, intermediate and advanced life support), to a two-tiered system. The latter consists of a two-year mid-level worker qualification and a four-year professional degree.
Out-of-hospital cardiac arrest in Johannesburg
The epidemiology of OHCA has never been studied in an African population. The only data dealing with OHCA outcomes in Africa were published in 2009, based on seven years’ accumulated information in an electronic clinical learning registry used by the Department of Emergency Medical Care at the University of Johannesburg. (Stein, 2009). Although not documenting every consecutive OHCA case within this time-frame, the study gives an impression of short-term OHCA survival and other characteristics in the Greater Johannesburg Metropolitan area, a region spanning roughly 1300 km2 with a population close to 8 million (Statistics South Africa, 2001)
Prehospital emergency care in the Greater Johannesburg Metropolitan Area is provided by several local authorities in the area acting as agents for the Gauteng Provincial Department of Health. The emergency services provided by these local authorities are all dual fire-EMS systems and use a two-tiered response to emergency cases. This involves the dispatch of ALS paramedics first, in non-transport primary response vehicles, followed by intermediate or basic life support personnel in ambulances. Several private EMSs also operate in the area, using the same two-tiered approach.
OHCA and peri-arrest related capabilities of ALS paramedics include defibrillation, endotracheal intubation and use of extraglottic airway devices, administration of intravenous adrenaline and other cardiac arrest-related drugs, synchronized cardioversion, external pacing and intraosseous drug administration in paediatric patients. In the absence of any special situations which may preclude this, ALS paramedics typically declare patients dead on scene upon termination of an unsuccessful resuscitation effort.
Results of the 2009 OHCA study in Johannesburg revealed three important differences when compared to most epidemiological OHCA studies:
It is likely that the latter two factors are related to the first.
Response times
The response times measured in this study only included the interval between emergency vehicle dispatch (i.e. receipt of dispatch information by paramedics in the vehicle) and arrival on the scene. Call centre time intervals were not available due to the difficulty in obtaining these data from a multitude of different local authorities and private EMS. Even using this shorter response interval, 48% of the cases responded to involved time intervals greater than eight minutes, with the greatest proportion in the eight to nine minute category and 22% of responses greater than 15 minutes in duration.
In contrast, many other studies show response intervals, measured as the time from call centre activation to arrival of an emergency vehicle on scene, between four (Finn et al, 2001) and nine (Huei-Ming, 2007) minutes in urban areas. It is unknown how much additional time call taking and dispatch would have added to the Johannesburg data, but at best this would have added several minutes to each response time.
Resuscitation
Only four out of every ten OHCA patients over the seven year study period in Johannesburg were resuscitated. The remainder were declared dead on scene without any resuscitative efforts. Consideration of recently published OHCA outcome studies shows that only one reported a smaller proportion of resuscitated patients. (Rewers et al, 2000). In contrast, a large OHCA outcome study in the USA involving 20 520 cases was associated with a 58% resuscitation attempt rate. The reasons for this low rate of attempted resuscitation in Johannesburg were not investigated as part of the study and remain unknown.However, it is possible that long response times may contribute to this effect as many paramedics may believe that resuscitation is futile after a 10 to 20 minute response.
Both the number of shockable initial rhythms and the number of patients with return of spontaneous circulation (ROSC) were low in the Johannesburg data. Of the resuscitated cases, 23% presented initially with a shockable rhythm and 18% had ROSC. These figures are lower than most others reported in the literature (Westfal et al. 1996; Kuisma and Maatta, 1996; Fischer et al. 1997; Rewers et al, 2000; Finn et al, 2001; Fairbanks et al, 2007; Grmec et al, 2007; Fridman et al, 2007). Further research is needed to investigate population and disease-related aspects of this pattern.
However, considering the time-sensitive nature of successful cardiac resuscitation (Weisfeld and Bekker, 2002), it is quite plausible that the observed proportions of non-shockable rhythms and low rate of ROSC is simply attributable to long intervals between patient collapse and paramedic arrival.
‘In addition to the competence of paramedics to adequately diagnose and resuscitate OHCA cases, the organization of EMS systems must allow these personnel to access patients rapidly enough to make a difference’
The contradictions of OHCA resuscitation in Johannesburg
Effective OHCA resuscitation that optimizes long-term survival is dependent on a combination of the right people at the right time for the OHCA victim. South African paramedic education is currently conducted in two streams: one involving in-service short course training and one involving university education.
Regardless of the type of qualification or educational programme related to it, OHCA resuscitation has always been a key feature. Learning outcomes and content are in line with the most recent scientific evidence disseminated by the Resuscitation Council of Southern Africa and assessment of OHCA resuscitation skills includes theoretical knowledge, objective structured clinical evaluations, simulated cases and assessment in the clinical environment. In all learning programmes, satisfactory performance in these areas is a requirement for qualification.
In addition to the competence of paramedics to adequately diagnose and resuscitate OHCA cases, the organization of EMS systems must allow these personnel to access patients rapidly enough to make a difference.
Data from the Johannesburg OHCA study indicated that, from an OHCA resuscitation perspective, most response times were unacceptably long. As mentioned previously, these response intervals did not include any call centre delays, only the interval between emergency vehicle dispatch and arrival. There are several possible reasons for the observed data.
Shortage of emergency vehicles
The most logical reason for long response times is a shortage of emergency vehicles. Inadequate vehicle numbers mean that each vehicle covers a large geographic area, with long distances to travel responding to each case and thus longer response times. Information on actual emergency vehicle numbers estimated for the Greater Johannesburg Metropolitan Area over the 2006-2007 period indicate that roughly half of these vehicles were not serviceable at any given time (Gauteng Department of Health, 2007).
This clear failure of systems management and oversight has significant implications for emergency vehicle availability and consequently for response times and OHCA outcomes.
Call centre delays
Although call centre delays were not described as part of the Johannesburg OHCA study data, it is very likely that such delays did occur and contribute to the small proportion of initial shockable rhythms observed.
The low emergency vehicle availability described above could logically be expected to have an aggravating effect on the timely dispatch of vehicles due to periods when all vehicles may be occupied and unavailable for dispatch. Additionally, the twotiered response system used throughout the Greater Johannesburg Metropolitan Area may impede the timely dispatch of primary response vehicles.
Two-tiered response systems
Two-tiered response systems, involving dispatch of a primary response vehicle followed by an ambulance, may work efficiently if there are an adequate number of ambulances available in a given area. If ambulance availability is low, the two-tiered response system may actually prolong response times overall by forcing primary response vehicles to wait on scene for longer periods until an ambulance is available to transport the patient to a receiving facility.
This ties up both ambulances and primary response vehicles for prolonged periods, forcing call centre dispatchers to delay dispatch of appropriate vehicles and paramedics to queued cases. Efficiency of the two-tiered response system is eroded by emergency services that do not have sufficient numbers of emergency vehicles available and it is quite likely that this impacts on response times in general and, by extension, OHCA outcomes.
Is OHCA research meaningful in an under-resourced setting?
Research in the area of OHCA resuscitation makes several basic assumptions, one of which is that patients will still be in a potentially viable state when resuscitated. Testing the therapeutic effectiveness of interventions when there are unlikely to be any survivors due to prolonged ‘down time’ is futile. The time-sensitive nature of OHCA resuscitation thus requires that meaningful research can only be carried out in systems guaranteeing the assumption of rapid patient access, thus precluding this kind of research in most under-resourced systems.
Conclusion
The term ‘under-resourced’ can refer to the lack of many different resources in general, or to specific resources. It is true that EMS in the Greater Johannesburg metropolitan area are not under-resourced in a general sense, but display a contradictory mix of resources; some adequate and others not.
At the highest levels, ALS care in South Africa involves paramedics with capabilities equal to those operating in some of the most advanced EMSs in the world. University degree programmes produce practitioners with well-rounded applied knowledge and an extensive skill set, including rapid sequence intubation and prehospital thrombolysis.
An external observer trying to gauge the impact of South African ALS paramedics on OHCA survival by examining the relevant educational programmes producing these practitioners would conclude that these outcomes should be excellent, based on the level of care and range of therapeutic options available.
Unfortunately, in many cases, the level of care of individual ALS paramedics is not equalled by the effectiveness or efficiency of the EMS systems in which they operate. Good OHCA outcomes appear to be less about highly qualified ALS paramedics and more about simply getting a defibrillator to the patient often enough, and in a time period short enough that it can be used effectively. The literature on resuscitation outcomes has, for some time, been emphasising the effectiveness of basic interventions (such as defibrillation and uninterrupted chest compressions) over more advanced ones for which there is little evidence of usefulness.
Consequently, EMS system flaws and resource short-falls which result in long response times are at odds with the first aim of OHCA resuscitation: accessing the patient as rapidly as possible.
Changing this situation is not easy. The educational landscape of prehospital emergency care is determined in South Africa by those working in the field themselves, in the form of the Professional Board for Emergency Care at the Health Professions Council of South Africa. Consequently, educational programmes for paramedics currently exist in a form that those in the profession desire. These programmes adequately address the needs of prehospital emergency care related to OHCA resuscitation and are in line with the latest available international evidence.
Management of EMS systems, however, is a function determined internally by the services themselves and is most certainly subject to many forces, often of an economic nature and often in direct competition to the need for enhanced resource allocation or structural system change. There is discordance between teachings and systems in prehospital emergency care, which produces emphasis in the incorrect place with regard to OHCA resuscitation in the Greater Johannesburg area.
Although the South African situation may not be characteristic of all under-resourced EMSs, the emphasis on building a sound EMS system before spending heavily on more skilled practitioners is of relevance anywhere, particularly in the quest to improve outcomes from OHCA.