This website is intended for healthcare professionals only.

Hospital Healthcare Europe
Hospital Healthcare Europe

Driving the development of a network of chest pain units across Germany

27 July, 2010  

The establishment of chest pain clinics is helping to reduce the time to treatment in the critical early stages of an acute coronary syndrome

Thomas Münzel
Department of Medicine II
University Medical Center
of the Johannes Gutenberg
University, Mainz

Germany is currently in the process of establishing a countrywide network of chest pain units (CPUs). In 2009, more than 50 CPUs were certified and within the next two years the expectation is that 200 more certified units will be assessing patients who present with chest pain.
The first CPU was established in Frankfurt in 2003, with Heidelberg and Mainz following suit soon after. From there, a certification scheme under the auspices of the German Society of Cardiology (GSC) was drawn up with a working group, which I chair. The Mainz clinic was set up in 2005, and the drive towards the growth of the network across the whole country is ongoing.
Initial evidence from Mainz CPU is showing benefits to patients and other hospitals and clinics within the catchment area, by offering a specialist and focused response to patients presenting with chest pain and also easing pressure on emergency departments (ED). Chest pain is one of the most common symptoms in ED,[1] comprising 5-20% of visits – although only 10-15% of patients with chest pain have acute coronary syndrome (ACS).
The idea to establish CPUs in Germany was primarily based on the success of chest pain centres (CPC) in the United States, where, by 2004, there were some 1,500 CPCs. Several studies comparing the classic emergency department, or an ED with a closely associated CPU, revealed that the prognosis of patients with acute chest pain may be better when these patients are treated in the CPC compared with the ED.[2] The establishment of CPCs in the US prevented the discharge of patients with an
ACS and was cost saving due to the prevention of complications in patients with an acute MI or unstable angina falsely discharged from the ED.[3]
In Germany, through the GSC certification process to ensure CPUs meet the necessary criteria, the goal has been to expand and regulate the network and to avoid the experiences of other
European countries that have been less successful in establishing CPUs. In the UK, progress in this area has been described as disorganised,[4] while in Spain – despite a recommendation in 2002 by the Cardiac Society of Spain to set up CPUs in all EDs to provide fast and efficient care for patients
with chest pain – just four had been established by 2008, suggesting that EDs in Europe are unenthusiastic about CPUs.[5]
The team at Mainz published several articles on the CPU’s early success. In a most recent publication, the team demonstrated an improved outcome for patients with an acute coronary
syndrome being treated in the CPU compared with the ED. Following that, I was asked by the GSC to establish a specification requirement for having a CPU in Germany.
To gain a certificate from the GSC, a CPU must have access to a catheter laboratory and four interventionalists, 24-hour  catheter laboratory readiness, four beds with monitors, be sited close to the ED and catheter laboratory, and have facilities to perform stress tests such as ergometry or stress echo.
A CPU is funded in the same way as an ED and is typically equipped with a 12-lead ECG, blood pressure measurement, transthoracic echo, transoesophageal echo, monitor for supervising heart rhythm, ventilation device, analysis for
laboratory value and biomarker of troponin I and T, external pacemaker and the ability to perform an exercise ECG or a stress echo after exclusion of acute MI.


In Germany, CPUs are a distinct structural unit with nurses, internists and cardiologists. They are located in private practices, city hospitals and university hospitals, and, ideally, they are positioned close to the ED and in the vicinity of a catheter laboratory. Patients are referred either by emergency physicians, private practitioners, cardiologists, or self-referral, which is becoming more important. The fundamental goal is a protocol-driven, systematic approach to the management of patients with chest pain with commonly used guidelines (Figure 1). These strategies allow a rapid initiation of a crucial therapy in high-risk patients with an acute coronary syndrome requiring admission and also the stratifi cation of a low-risk group of patients who can be safely discharged and managed as outpatients.
The primary goal of establishing a CPU is to reduce the time it takes for a patient experiencing symptoms of a possible heart attack to see a physician, thus reducing the time to treatment
in the critical early stages of an acute coronary syndrome. Another important aspect is to provide a specialised unit where doctors can better monitor patients when it is not clear
whether they have myocardial ischaemia or not. With these measures, one hopes to ensure that one can prevent a patient being sent home too early or needlessly occupying a coronary care unit bed.
Patients presenting at a CPU with acute chest pain and  ST-segment elevation are admitted to the catheter laboratory within 90 minutes without further diagnosis in the CPU. Those
without ST-elevation are monitored and their risk factors assessed, and they are either admitted to the catheter lab where necessary or discharged. On average a patient in this category can be discharged within six hours. It is important to fi nd out who may develop a myocardial infarction and who will not. About 55% of all patients admitted to the CPU can be discharged – a number that is very similar to data reported from the US. From the 45% of the patients not being discharged, 50% are transferred for a diagnostic cath and about 50% ultimately get a percutaneous coronary intervention.
Chest pain that requires admission to the CPU may be typical or atypical. It is of utmost importance to inform people about atypical chest discomfort, which also may be an early sign of
myocardial ischaemia, to prevent time delays. In Germany, it is common for patients to wait until the discomfort becomes painful enough to call the emergency 112 number. It is almost as if the chest pain has to stop one dead in one’s tracks
and this is actually what happens. This is why 70,000 Germans each year lose their lives, and many of these deaths are clearly preventable.
An important part of a CPU’s success is marketing and raising public awareness of the availability of such a facility and the risks to people of not reacting quickly enough to chest pain. Cardiologists were discovering that the threshold before a patient called the emergency number was high, particularly if their symptoms were not typical of an acute MI.


 In Mainz, the CPU has been working with the city’s soccer team, Mainz 05, for the past three years to boost an awareness campaign about chest pain and the CPU. They have put posters on buses and trams and hung a banner at the main station in Mainz (Figure 2). The concept follows the lead of US chest pain centres that have enlisted the support of Major League Baseball teams, which have adopted the Early Heart Attack Care (EHAC) awareness programme. The growth of CPUs in Germany is now
moving ahead rapidly. At present, 50 CPUs have received GSC certification and a further 100 have applied for it. By the end of 2010, it is expected there will be about 200. With myself heading the working group responsible for the certification
process, 60 cardiologists currently perform the audits in clinics applying for certification. The CPU in Mainz has shown early benefits to patients and is proving cost-effective, reflecting
findings in the US[3] and UK.[6]
The situation in Mainz is unique because all the city’s hospitals, private cardiologists and private practitioners acknowledge that if a patient has chest pain, the only place to be assessed is the CPU at the university hospital. That has been a factor in the number of patients growing steadily from 751 in 2005 to 1,960 in 2008. An important advantage of a CPU is that patients
in whom a myocardial infarction is excluded due to negative biomarker screening (negative echo or lack of ECG changes), also get an exercise test and are discharged with a complete diagnostic workout so that the practitioner does not have to
perform additional diagnostic tests. An analysis of patients at Mainz CPU also demonstrates that the average time before
discharge of a patient with an ACS in a CPU is shorter than that of a patient with the same diagnosis treated in the ED (6.4 hours compared to 11.6 hours). Similarly, for patients receiving
a catheter, the average time until discharge dropped substantially from 3.2 days to 2.4 days.  Published research on the impact of the clinics is rare at present, although a study
from the Mainz clinic about the improvement of sensitivity and specificity for detecting a myocardial infarction with an ultra-sensitive troponin I assay has been accepted in the New
England Journal of Medicine. [7]
Another manuscript, currently in the review process, analysed 1,796 patients, with 483 suffering ACS. It demonstrated significant benefit in long-term outcome with regards to death, myocardial infarction or stroke within a year in patients treated for ACS in the chest pain unit compared with patients treated in the ED. It shows, for the first time, that implementation of a
chest pain unit significantly improves patient care (Figure 3) with regards to long-term outcome, in addition to  cost-effectiveness.[8] Within the next two years the GSC will certify between 150 and 200 CPUs and, eventually, the aim is to establish a network of 300 that the working group believes will be adequate to cover the whole of Germany.
But a wider goal is now emerging, based on the success of the CPU programme in Germany – to encourage the structured establishment of CPUs across other regions of Europe.

1. Barish RA et al. J Healthc Qual 1997;19:6-12.
2. Amsterdam EA et al. Cardiol Clin 2002;20:117-36.
3. Farkouh ME et al. N Engl J Med 1998;339:1882-8.
4. Cross E et al. Emerg Med J 2007;24:100-2.
5. Miro O et al. Emerg Med J 2008;25:59-60.
6. Goodacre S et al. Emerg Med J 2003;20:429-33.
7. Keller T et al. N Engl J Med 2009;361:868-77.
8. Keller T et al. Clin Res Cardiol 2010; 99:149-55.