Syncope Unit: rationale and requirement – the European Heart Rhythm Association position statement endorsed by the Heart Rhythm Society

External contributors to the Task Force: Mehran Asgari (UK), Gonzalo Baron-Esquivias (Spain), Jean-Jacques Blanc (France), Ivo Casagranda (Italy), Conal Cunnigham (Ireland), Artur Fedorowski (Sweden), Raffello Furlan (Italy), Nicholas Gall (UK), Frederik J. De Lange (The Netherlands), Geraldine Mcmahon (Ireland), Peter Mitro (Slovakia), Artur Pietrucha (Poland), Cristian Podoleanu (Romania), Antonio Raviele (Italy), David Benditt (USA), Andrew Krahn (Canada), Carlos Arturo Morillo (Canada), Brian Olshansky (USA), Satish Raj (USA), Robert Sheldon (Canada), Win Kuang Shen (USA), Benjamin Sun (USA), Denise Hachul (Brazil), Haruhiko Abe (Japan), and Toshyuki Furukawa (Japan)


Scope of the document
The European Society of Cardiology (ESC) has played an important role in advancing our understanding of the causes, optimal investigation, and management of syncope through publication of practice guidelines in 2001, 2004, and 2009. 1 -3 The 2009 ESC guidelines recommend the establishment of formal Syncope Units (SUs)-either virtual or physical site within a hospital or clinic facility-with access to syncope specialists and specialized equipment. 3 In response, this position statement by the European Heart Rhythm Association (EHRA) endorsed by the Heart Rhythm Society (HRS) offers a pragmatic approach to the rationale and requirement for an SU, based on specialist consensus, existing practice and scientific evidence (see Appendix).
The panel consists of specialists who have experience in developing and leading such units representing cardiology, geriatric and general internal medicine, neurology, and emergency medicine.
This document is addressed to physicians and others in administration, who are interested in establishing an SU in their hospital, so that they can meet the standards proposed by ESC-EHRA-HRS. 1 -3

Definitions Definition of syncope and transient loss of consciousness
Syncope is a transient loss of consciousness (T-LOC) due to transient global cerebral hypoperfusion, and is characterized by rapid onset, short duration, and spontaneous complete recovery. This definition of syncope has been developed by the Task Force for the Diagnosis and Management of Syncope of the ESC and endorsed by the EHRA, European Heart Failure Association and European Heart Rhythm Society. Transient loss of consciousness is a term that encompasses all disorders characterized by self-limited LOC, irrespective of mechanism. 1 -3 By including the mechanism of unconsciousness, i.e. transient global cerebral hypoperfusion, the current syncope definition excludes other causes of T-LOC such as epileptic seizures and concussion, as well as certain common syncope mimics, such as psychogenic pseudosyncope.

Definition of a Syncope Unit
An SU is a facility featuring a standardized approach to the diagnosis and management of T-LOC and related symptoms, with dedicated staff and access to appropriate diagnostics and therapies. The SU should also take the lead in educating and training clinicians who encounter syncope. Even if the most appropriate term describing such an organization should be the more general T-LOC Unit (or Faint Unit), this Task Force decided to maintain the term of SU, because it is most frequently used worldwide. This Position Paper is a pragmatic approach to outline the constituents of an SU and assist target groups with the current available necessary information. The authors emphasize that there is, at present, insufficient available evidence whether an SU (examples of a number of models are detailed later in the document) is superior in efficiency or outcomes to a syncope specialist 4 or newer technologically driven models of syncope management. 5 We anticipate that the Position Paper will stimulate structured research to determine best practice models for T-LOC evaluation in different settings and cultures.

Rationale for a Syncope Unit
Expected benefit and barriers to setting-up a Syncope Unit Syncope is a common medical problem that can be debilitating and associated with high healthcare costs. 6 -9 There is wide variation in practice of syncope evaluation, and wide variation in adoption of recommendations from published guidelines. 10,11 The absence of a systematic approach to T-LOC incurs higher health and social care costs, unnecessary hospitalizations, and diagnostic procedures, prolongation of hospital stays, lower diagnostic rates, and higher rates of symptom recurrences. Therefore, a systematic approach, by a dedicated service (an SU), equipped to evaluate and manage this common problem may ensure better management of T-LOC, from risk stratification to diagnosis, therapy and follow-up ( Table 1).
Despite the recommendation from the ESC, 2,3 SUs are not widely established in clinical practice. Possible reasons for this are outlined in Table 2. Barriers to establishing an SU include lack of resources, lack of trained dedicated staff, and complex presentations to multiple settings, necessitating involvement from multiple disciplines. When developing a case of need for the SU, individual practices may not be able to access detailed information to inform fully the economic cost and resource requirements necessary and this can make the justification for realignment of resources challenging. This document will assist practitioners to develop a model best suited to local requirements.

Syncope Unit reduces underdiagnosis and misdiagnosis of syncope
An Internet search of the phrase 'Misdiagnosis of Syncope', reveals about equal numbers of search hits from three perspectives: those who approach the problem from the perspective of the over diagnosis of epilepsy, 28 the underdiagnosis of syncope, 29 or legal firms soliciting business from victims of either. When it comes to the underdiagnosis and misdiagnosis of syncope, are estimated to be as high as 40%. 17,28 -32 Underdiagnosis and misdiagnosis have been reported in both outpatient and emergency settings. Although there are no large-scale randomized trials comparing misdiagnosis of syncope in SUs to usual care, smaller cohort studies confirm high rates of misdiagnosis with usual care and the benefits of a structured approach to diagnosis. 9,17,20,33 Syncope Unit reduces hospitalization In hospitals without an SU, T-LOC evaluation and management is more frequently carried out as an inpatient rather than an outpatient service. In other words, patients are preferentially admitted to emergency services rather than evaluated and managed as outpatients ( Table 3). In one study, the average length of stay for acute admissions due to syncope or collapse was two-fold higher in hospitals without an SU. 12,15,34,35 The Syncope Evaluation in the Emergency Department Study (SEEDS) 15 randomized intermediate-risk patients to an Emergency Department (ED)-based explicit syncope protocol vs. routine inpatient admission. Hospital admissions were reduced by 56%, and total patient-hospital days were reduced by 54%. In the Emergency Department Observation Syncope Protocol (EDOSP) trial, 26 patients randomized to an ED observation protocol experienced a 77% reduction in hospitalization and a 40% reduction in hospital length of stay. 26 An integrated model of Short Observation stay in ED, coupled with fast track to an SU allowed a reduction of the admission rate to 29% with 20% of patients being dicharged after a short observation in ED, 20% fast-tracked to the SU and 31% directly discharged. 36 Syncope Unit reduces cost of syncope Cost estimates Several studies investigated costs of syncope as they appear in national or hospital records (Table 3). 24,34,37,38 In the USA, in 2004, the mean cost for a syncope-related admission was $5400 (95% CI: $5100-5600) with a total annual cost of $2.4 billion, 32 similar to asthma ($2.8 billion) and HIV ($2.2 billion). In Italy, in 2006, 17 the mean cost for a syncope-related admission was E2785 + 2168; hospital costs accounted for about three quarters of total costs. The cost per patient discharged from ED was E180 + 63. In Spain, in 2006, 9 the overall cost (which included stay, diagnosis, and treatment) per admitted patient was E11 158 (range: 1651-31 762). In the outpatient setting, the cost is similarly alarmingly high due to significant variability in practice and the use of unnecessary tests. 6,19,39 In one study, 40 patients with unexplained syncope had a median of 13 nondiagnostic tests performed (range: 9 -20) before receiving an implantable loop recorder.

Cost reductions
Various hospitals organized syncope care through the creation of SUs, where a solid conceptual framework with clearly delineated diagnostic procedures is implemented. The primary outcome was an increase in the rate of diagnostic yield, 1 -3 and a reduction in costs primarily by reducing the number of admissions, duration of hospital stay, and the number of unnecessary tests, 12,13,15,17,19,21 -23,25,41,42 with few exceptions. 14 Indeed, Brignole et al. 17 have shown a 19% reduction in cost per patient and a 29% reduction in cost per diagnosis in the standardized care group when compared with the conventional approach. The EDOSP randomized trial 26 included an explicit cost analysis. Hospital facility costs were $629 less in the observation unit group compared with the routine admission group. Rates of diagnostic testing and specialty consultation were similar in both groups; therefore, cost savings were due to reduction in hospital length-of-stay.

Expected benefit
Establishing an SU should benefit three parties: (i) patients by increasing the rate of correct diagnosis, (ii) health payers by reducing total cost per patient and diagnosis, and (iii) hospitals by increasing their market share via the added value proposition. In healthcare systems based on payment per test or per medical action, an SU may reduce income through a reduction of tests.  There may be cost savings, but these will be system dependent and thus vary. The cost benefit of a syncope specialist or an SU in different settings and different healthcare systems has not been exposed to rigorous economic and scientific scrutiny. Further research is required to determine resource outcomes and the authors acknowledge the limitations of the current knowledge base and recognize that service models may be influenced by local circumstances.

Structure of European Heart Rhythm Association Syncope Unit
Existing models for syncope management Syncope management organization may differ widely among healthcare systems and from hospital to hospital. A review of published data on organization and impact on outcomes of models of care may guide SU design and implementation for a given environment. Table 4 summarizes these data. We acknowledge that overlap exists between these models.
A functional Syncope Unit located in a cardiology department 13,17,20,25,39,43 -45 In this model, introduced in Italy and adopted in other countries such as Sweden, Portugal, USA, and France, the SU is supervised by cardiologists, supported by dedicated personnel with expertise in syncope. Patients access the SU mainly as outpatient or via the emergency room. This model has evolved into a 'virtual' unit based on the expertise of a limited team and in some instances on webbased decision-making software. 52 Access to a specialist is regarded as essential. A specialist can be accessed by any means, e.g. telephone.
The Evaluation of Guidelines in SYncope Study-2 (EGSYS-2) 17 showed a sharp decrease in the overall cost of care driven by a reduction in average cost per patient of 19% and average cost per diagnosis of 29%. Seventy-one Italian hospitals now have an SU that has been certified by peer-review members of GIMSI (Gruppo Italiano Multidisciplinare per lo studio della Sincope, www.gimsi.it). Similar models have been described in other departments, e.g. geriatric and internal medicine; the organization is basically the same with formalized fast-tracking processes to cardiological testing. In a few instances, the SU includes also a short observation stay as part of an internal protocol for risk stratification of intermediate risk ED patients. 36 The Day-Care Syncope Evaluation Unit and the Falls and Syncope Services This model was first developed in Newcastle, UK 12 and takes the form of an outpatient, day-care facility located in a general hospital.
The service provides a multidisciplinary approach based on the application of evidence-based diagnostic algorithms to patients with falls (for older patients) and T-LOC of suspected syncopal nature (all adult age groups). Close liaison exists with acute medical in-patient and ED services. After a consultation with an emergency physician, geriatrician, internist, or general practitioner, patients have access to non-invasive diagnostic testing, occupational and physiotherapy, and supplementary examinations supported by specialist nurses. There is close cooperation and consultation with cardiology and neurology services for further diagnostics and treatments. In line with this experience, Falls and Syncope Services for older people have evolved by defining protocols and educational methods for inpatients 35 and outpatients. 46 The Rapid Access Blackouts Clinic This type of SU functions as a 'referral centre' for patients with T-LOC. 47,48 As such, it is positioned between first response and specialist referrals. It is led jointly by a cardiologist and a neurologist. The aim is to provide rapid access to clinical and ECG assessment in order to screen patients. The SU is run by nurses specialized in epilepsy, cardiac arrhythmias, or geriatrics. Patients referred by general practitioners or emergency services complete a triage questionnaire: 60 standard questions/data detailing characteristics of falls, syncope, and epilepsy, which is analysed by the nurses. A cardiologist may be consulted for interpretation of tests, notably the ECG. Following this evaluation, patients are referred, as appropriate, to a cardiologist, neurologist, geriatrician, general practitioner, or psychologist. Continuity of care is ensured by maintaining and sharing a database for all stakeholders.
Tertiary referral Syncope Unit 49,53 This model is centred on one syncope specialist, a neurologist, internist, or cardiologist, who mostly sees tertiary T-LOC referrals from neurologists and cardiologists. The SU consists of an outpatient clinic and a core laboratory performing tilt table test, carotid sinus massage, cardiovascular autonomic tests, and ambulatory BP monitoring. The tertiary character means that other ancillary tests have already been performed, and that the case mix concerns low rates of cardiac syncope and of unexplained syncope, but high rates of reflex syncope, psychogenic pseudosyncope, and epilepsy (see 'Competence' section).

The Syncope Observational Unit in the Emergency Department
This type of SU is described in the SEEDS study. 15 This single-centre prospective randomized study evaluated a standardized unit incorporated in the ED of a university tertiary care hospital, compared with conventional care, in a group of syncope patients considered to be at 'intermediate' cardiovascular risk. After 6 h of monitoring and tests in the ED which included regular orthostatic blood pressure measurements, tilt table test and carotid sinus massage upon physician's request and electrophysiologist's consultation when requested, patients without indication for hospitalization were offered rapid outpatient follow-up consultation. This study showed, in 51 consecutive patients randomized to SU evaluation, an improvement in diagnostic yield compared with conventional care (67 vs. 10%) and a decrease in hospital admission (43 vs. 98%), but no changes in the average length of stay. The model did not reduce 2-year mortality nor syncopal recurrences. Using a protocol with an initial evaluation similar to that described in the ESC Guidelines and an observational unit, for up to 24 h, within the ED, a Spanish group has achieved a diagnosis in 78% of patients  presenting as emergencies with only 10% of syncope patients being admitted to hospital in this model. 51 Recently, the EDOSP study 26 evaluated an ED observation protocol at five sites, including university, community, and public hospitals. Patients at 'intermediate' risk were randomized either to an explicit ED-based observational unit protocol vs. routine care. The observational unit protocol included up to 24 h of cardiac monitoring and echocardiogram for selected patients. In 124 randomized patients, there were reductions in hospitalizations rate (15 vs. 92%), length-of-stay (29 vs. 47 h), and hospital costs were $629 lower than the admission group. There were no differences in safety events (i.e. serious 30-day outcomes occurring after hospital discharge), quality-of-life, or costs. The EDOSP study generalizes the SEEDS findings to a diverse set of hospitals, and includes novel assessments of patient-centred outcomes and a formal economic analysis.

The Faint and Fall Clinic
The Faint and Fall Clinic offers a multidisciplinary approach to patients presenting with fainting spells or falls. Patients are evaluated by advanced nurse practitioners and then seen by a cardiologist or a geriatrician with rapid access to a neurologist, as needed. Providers in the clinic use a web-based interactive software that integrates the most recent guidelines for risk assessment and management of patients with T-LOC. 5

Current situation
The UK has a growing number of SUs, which are listed by STARS (Syncope Trust and Anoxic Reflex Seizures), a charitable organization providing information to patients about syncope and related conditions, www.stars.org. Italy has a growing number of SUs which are listed by GIMSI (Gruppo Italiano Multidisciplinare per lo studio della Sincope), www.gimsi.it. These information sites include available SUs that are geographically close to the enquiring patient. In Ireland, USA, Canada, Spain, Portugal, France, Netherlands, Sweden, Japan, Brazil, Romania, Poland, Slovakia, and other countries there are similar developments.
In summary, whatever the SU model, the key elements are rapid access to syncope expertise in trained, dedicated staff, together with the utilization of standardized algorithms. The European Heart Rhythm Association considers that SUs should be widely available in Europe. Their aims and structure should be in line with one or other of the models reported permits each hospital to develop their own model to suit its particular environment.

General attributes of a Syncope Unit
The SU can be virtual or based on a pre-defined location such as a unit associated with the ED, an ambulatory clinic, or employs a combination of approaches. The model of SU should be the best fit for local practice. Because of the extensive differential diagnoses and high prevalence of syncope in older patients, the skill mix of SU staff should include some training/knowledge of common disorders that cause or mimic syncope: commonly cardiovascular, neurologic, geriatric, and psychiatric disorders.
As no single syncope care service model is suitable for all healthcare systems, the following is a list of some of the important features to consider when establishing an SU: † Structure of SU W The model of care delivery should be appropriate to local resources and local specialities while ensuring implementation of published practice guidelines.
W Models of care delivery vary from a single one-site-one-stop syncope facility to a wider multifaceted practice in which several specialists are involved in syncope management.
W The SU can be a single site facility or virtual model with mobile team. † Stakeholders W All key stakeholders should be involved in the earliest stages of development and implementation of the SU. † It is essential to establish a mechanism through which regular communication can be established with all stakeholders (i.e. patients, referring physicians, hospital/clinic management, consultant physicians, nurses, and other allied medical professionals) in order to ensure an ongoing consensus for and understanding of proposed management strategies. This mechanism includes the implications of and implementation of published guidelines. Stakeholders may be staff from the ED, neurology, general internal medicine, orthopaedic surgery, geriatric medicine, psychiatry, and ear, nose, and throat (ENT), paediatric departments in addition to cardiology. † This should also include agreed measurement metrics of performance in order that early recognition of variation is achieved; thereby re-calibration of the model can take place in a timely fashion. These measures should include outcome measures, process measures, and balancing measures. † A clear diagnostic and therapeutic pathway provides a framework, which is fundamental to enable new evidence to be incorporated into the model seamlessly. † Management W The management strategy should be agreed on and practiced by all practitioners (encompassing a range of specialities) involved in syncope management. † Patient case mix W The age range and symptom characteristics of patients appropriate for syncope investigation should be determined in advance. Some facilities are prepared to evaluate both paediatric and adult patients, whereas others limit practice to adult or paediatric cases. A wide age range is encouraged. † Referral sources W Potential referral sources should be taken into consideration.
Referral can be directly from family practitioners, from the ED, from occupational physicians, from hospital admissions, and from patients in institutional settings. The scope of referral source has implications for resources and skill mix. † SU-skill mix and staffing W There are existing models in which cardiologists (commonly with an interest in cardiac pacing and electrophysiology), neurologists (commonly with an interest in autonomic disorders and/or epilepsy), internists (commonly with an interest in cardiovascular physiology and autonomic disorders), emergency doctors, and geriatricians (commonly with an interest in age-related cardiology or falls) each may lead syncope facilities. There is no evidence for superiority of any model. W The skill mix (i.e. the types of professional/specialities) required to staff the facility depends on the extent to which screening of referrals occurs before presentation at the facility. For example, if referrals are directly from the community a broader skill mix than cardiology is required. Under these circumstances, other disorders such as epilepsy, autonomic disorders, neurodegenerative disorders, metabolic disorders, and falls are also common.
W When establishing a unit the lead clinician should have knowledge of the catchment area for referrals and projected volume in order to estimate staff requirements and to tailor the scope of referrals to available resources. The volume of activity and the number of personnel largely vary based on the model of SU and local organization. No empirical figures can therefore be given. However, since the aim of this document is to provide practical advice to stakeholders who are interested in setting-up an SU, as a general guide, this Task Force believes that the following figures should be of help. In the Italian experience, 41 163 patients per 100 000 inhabitants per year were referred to the local SU. They performed an average of 2.9 tests per patient. Patients will be followed-up from a minimum of one to multiple visits. Multiple visits (on-site or by means of telemedicine) were necessary especially in the case of patients with unexplained syncope undergoing prolonged monitoring or patients who had received device therapies (pacemaker clinic, etc.). Thus, this Task Force estimates that one syncope specialist and one technician need to work the equivalent of one full working day per week for every 100 000 inhabitants of the catchment area.
W SU staff should provide ongoing education and training in syncope diagnosis, investigation and management to primary and secondary care colleagues who deal with this symptom in their day-to-day practice.

Structure of the European Heart Rhythm Association Syncope Unit
The proposed structure of the EHRA SU is shown in the Consensus Statement 1. The role of physician and staff in performing procedures and tests is shown in Table 5.  Current practice limited to few countries. c Biofeedback means that the training session of the counter pressure manoeuvres consists of biofeedback training using a continuous blood pressure monitor. Each manoeuvre is demonstrated and explained. The manoeuvres are practiced under supervision, with immediate feedback of the recordings to gain optimal performance. 54 Consensus Statement 1-Structure of the EHRA SU Staffing of an SU is composed of: (1) One or more physicians of any specialty who are syncope specialists. Owing to the multidisciplinary nature of T-LOC management, each SU should identify specific specialists for SU and for consultancies. (2) A team comprising professionals who will advance the care of syncope patients. These may be physicians, specialized/trained nurses, or others who bring multidisciplinary skills to the facility, coupled with administrative support. The roles played by members of the team may vary according to local circumstances and individual skill. Nurses may be expected to take very important roles including initial evaluation of patients, follow-up clinic assessments, selection of investigations including tilt testing and implantation/insertion of ECG loop recorders according to pre-defined protocols and local regulations. (3) Given that the SU is integrated within a hospital organization, syncope specialists, and staff are not necessarily employed full-time, but frequently have other duties depending on the volume of activity in the SU.
Facility, protocol, and equipment (1) An SU will deliver most of its care to outpatients in addition to ED and inpatients (2) The SU should follow an internal protocol, which applies to diagnosis and management and is agreed by stakeholders (3) An equipped facility must be available (4) Essential equipment/tests: -12-Lead ECG and 3-lead ECG monitoring -Non-invasive beat-to-beat blood pressure monitor with recording facilities for subsequent analysis -Tilt The 'syncope specialist' The syncope specialist has responsibility for the comprehensive management of the patient from risk stratification to diagnosis, therapy, and follow-up, through a standardized protocol. The syncope specialist requires specific knowledge. The domains required are listed in the 'Competence' section.

The staff
Most of the work is undertaken by nursing/technical staff. This requires specific skill and competence. In addition to assisting the syncope specialist, the specialized nurse/technician will perform procedures and tests (under physician supervision) provided that they are based on internal protocols and rules ( Table 5).

Competence Considerations
Defining the area and level of competence of an SU is based on the following assumptions. Even if the skill of an individual syncope specialist may be insuffcient to cover the whole case mix of the SU, the multidisciplinary skill of the different specialists involved in the SU should potentially be competent in are all disorders referred to as T-LOC. At present, the field lacks structured accreditation for clinical skills of the clinicians, and additional staff as well as equipment and facilities in the SU. The authors anticipate that this Position Paper will stimulate new structured training and accreditation opportunities. A diploma course, 'Syncope and Related disorders' for international participants, awarded by the Royal College of Physicians in Dublin, Ireland, is one example (www.rcpi.ie).
A second factor determining competence is risk management: the risks of cardiac syncope are high, those of reflex syncope low, with epilepsy, orthostatic hypotension and psychogenic T-LOC being intermediate. High risks require higher diagnostic skill levels regardless of frequency.
Thirdly, patient age affects the scope of competence. Syncope Units focusing on the elderly will encounter a different set of disorders compared with those with paediatric patients. 46 While paediatric T-LOC/syncope is not covered in depth by the 2009 ESC guidelines, the basic approach is the same as in adults, and children with T-LOC may profit as much from an SU as adults.

Syncope specialist
The considerations above prompted the following pragmatic description of a syncope specialist. A syncope specialist is a physician who has sufficient knowledge of historical clues and physical findings to recognize all major T-LOC forms, including mimics, as well as syndromes of orthostatic intolerance.
Syncope specialists need not all have the same skill levels, but the SU as a whole must be able to provide a minimum skill set, so a combination of specialty skills is optimal. These conclusions are specified in Consensus Statement 2-Competence and skills mix of physicians and staff required for syncope management in an SU.

Notes on training
Syncope specialists typically start working in an SU after specialty training, so knowledge regarding T-LOC forms not covered by their specialty may need to be refreshed. Reflex syncope, orthostatic intolerance, and psychogenic pseudosyncope deserve special attention, as they usually do not routinely feature in prior training.
Reflex syncope is frequent in the population (30 -40%) and in SUs (59%). Mastering the diagnosis of reflex syncope is difficult because its signs and symptoms are so variable that syncope cannot be defined practically using clinical descriptors: the ESC Guidelines defined T-LOC clinically but not syncope. 57 No ancillary test for reflex syncope meets the requirements, 58 leaving history taking as the prime diagnostic instrument. The importance of history taking and its high diagnostic yield 49 means that history taking should be allowed time. Syncope specialists typically set aside more time for history taking than novices, and may need up to 60 min to take a history and explain a diagnosis. 58 -62 As for risk prediction, rule sets did not perform better than clinical judgement. 63 A thorough knowledge of circulatory physiology helps to attribute historical elements to known circulatory patterns, strengthened by tilt table test experience. 64,65 Psychogenic pseudosyncope occurs at vastly different frequencies in different studies. It can be recognized through history taking and often with tilt table testing. 66 Video-EEG monitoring is a preferable addition, conforming to the gold standard approach for psychogenic non-epileptic seizures.

Legend
The levels described here concern the SU as a whole, not those of individual physicians, except for column 1: the requested level of minimum basic diagnostic skills applies to each syncope specialist.

Diagnostic skills per syncope specialist
These skills refer to history taking and physical examination only. 'General medical knowledge': this is the level stipulated by medical education, i.e. the knowledge medical students are required to have in order to pass medical examinations.'Specialist knowledge': this is the level required for specialty examination; when no specialty demands are available as for reflex syncope, the specialist level denotes the ability to recognize rare forms occurring once in about 200 cases.

SU access to ancillary tests
Ancillary diagnostic skills concern all necessary diagnostic tests. Full access: the SU carries out the test. Preferential access : the SU does not usually carry out the test but has preferential access to the tests/services in question.

SU ancillary tests
This is an overview of the most likely test required for the disorder in question.

Comprehensive care
The possibility of providing a comprehensive standardized syncope management of the patient from risk stratification to diagnosis, therapy, and follow-up characterizes and distinguishes an SU ( Figure 1).

Referrals, access criteria
Referral can be directly from family practitioners, ED, in-hospital and out-hospital services, or self-referral from the patient.  (so-called 'protected discharge' or 'advanced access' with an appointment for early assessment), in order to reduce hospitalization rates, directly from the ED or after a short stay in the Short Observation unit of the ED as part of an internal protocol for risk stratification of intermediate risk ED patients. 36 In an existing multicentre experience, 41 the majority of the patients (60%) were referred from out-of-hospital services, 11% had immediate referral and 13% delayed referral from the ED, and 16% were hospitalized. One of the critical variables for referrals from EDs is the initial clinical assessment by the non-specialist in syncope regarding the decision to admit. Since this issue is outside the scope of the present document, we refer to the existing guidelines, such as for example the ESC guidelines 3 and the Canadian Specialist Consensus 4 for defined algorithms on risk stratification and aligned actions regarding admission criteria.

Consensus Statement 2-Competence and skills mix of physicians and staff required for syncope management in an SU
Who should access the SU? † Patients affected by T-LOC of suspected syncopal nature who, because of frequency or severity of the episodes, need to establish a diagnosis and to ascertain the need for a specific therapy. † Patients with an already established diagnosis in order to receive a specialist's consultancy on the best evidence-based therapy or to start specific treatment, i.e. those listed in Consensus Statement 4. † Patients who need follow-up to make a final diagnosis or assess efficacy of therapy (see 'Final report and follow-up' section) Who should not access the SU? † Patients with a certain diagnosis and/or an established indication for therapy, e.g. patients with bradycardia with guideline-based indications for cardiac pacing. † Patients in whom syncope is a symptom secondary to underlying disease which requires urgent and specific diagnostic and therapeutic pathways that cannot be followed within an SU, e.g. syncope due to acute myocardial infarction or acute bleeding.
Only a minority of syncope patients seeking medical advice needs to be referred to an SU. In an existing multicentre experience, 41 adult patients referred to an SU accounted for 163 per 100 000 inhabitants per year. The corresponding yearly incidence of syncope per 100 000 in the general population, ranges from 950 per 100 000 of the overall population of the Utah study 67 to 1100 per 100 000 in the patients .40 years old in the Framingham study. 68 By comparison, the yearly prevalence for patient visits to general practitioners is 930 per 100 000 inhabitants in a study performed in the Netherlands 6 and 260 per 100 000 for patients referred to an ED in a multicentre Italian study. 17

Management modalities
Diagnostic pathway Diagnostic assessment consists of core assessment directly performed by the personnel of the SU and, when appropriate, by a fast-track access to any other tests and specialist visits that are not directly performed in the SU.

Treatment pathway
Even if the model of therapy delivery is that which is most appropriate to existing practice and resources, those listed in Consensus Statement 4 are the minimum core treatments that should be provided directly by the SU staff.

Final report and follow-up
Once the work-up is complete and a therapy is prescribed, the patient and referring practitioner should receive a written comprehensive final report and be referred back to the referring physician. The framework of a structured final report is described in Tables 6 and 7. However, the patient should still be periodically followed-up by the SU staff for three main purposes: 1. to assess the efficacy of treatment and re-evaluation in the case of syncope recurrence, 2. to establish a diagnosis in those patients with unexplained syncope who are undergoing prolonged ECG monitoring with external or implantable loop recorders, and 3. to contribute to the program of quality control of the SU (followup database, audit, etc.).
The follow-up process can be easily performed via phone interview or other tools, e.g. web-based.

Quality indicators
Based on the data presented in the previous sections, this Task Force believes that reasonable goals (expected benefits) of the EHRA SU should be those listed in the Consensus Statement 5.
Since clinical outcomes will vary according to the local patient cohorts in different geographic locations, the method of comparative effectiveness to assess the relevant clinical outcomes before and after the implementation of the SU has been considered. The minimum acceptable values should be compared with usual care without SU. Published information to serve as baseline data may be used if local information is not available (e.g. national data, or hospital practice before establishing the SU, or practice with similar characteristics in nearby hospitals without SU). As the practice of the SU evolves, new outcome metrics will be developed; the below suggested metrics are recommendation for a starting SU. Alternative approaches may include reduction in unnecessary/ inappropriate investigations; the ability of individual practices to achieve these targets will depend on data available and local practices. Whereas the guidelines are aiming to standardize syncope management, this TF acknowledge that it may take some years to achieve this depending on maturity of services and local practice. If the episode of T-LOC was initially attributed to a syncopal condition but the subsequent evaluation demonstrated a non-syncopal mechanism (i.e. epilepsy, vertebro-basilar ischaemic attack, unexplained fall, or psychogenic [functional] pseudosyncope), hypoxia, intoxication, cataplexy, drop attacks) Unexplained syncope/T-LOC In those patients without any of the above diagnosis Syncope Unit: rationale and requirement the metrics are not intended to be imperative for practice but rather a rough guide and dependent on extant practice, adequate human and other resources and the specific healthcare setting.
These are advisory metrics and should not be adopted as mandatory targets until further research on preferred syncope management structures is conducted relevant to different countries and healthcare systems. Furthermore, the current field lacks structured accreditation programmes for T-LOC/Syncope and, until these are established and widely available, specified training metrics are difficult to implement. The details of these metrics should be reviewed and approved by the lead physician, the syncope team, the hospital clinical practice committee, and stakeholders. External or internal audit via a registry mechanism may be considered to evaluate the quality outcomes.