Article Text

Download PDFPDF

Original article
Safety and efficacy of paramedic treatment of regular supraventricular tachycardia: a randomised controlled trial
  1. S Honarbakhsh1,
  2. V Baker1,
  3. C Kirkby1,
  4. K Patel1,
  5. G Robinson1,
  6. S Antoniou1,
  7. L Richmond1,
  8. W Ullah1,
  9. R J Hunter1,
  10. M Finlay1,
  11. M J Earley1,
  12. M Whitbread2,
  13. R J Schilling1
  1. 1Cardiovascular Biomedical Research Unit, Barts Heart Centre, St Bartholomew's Hospital, Barts Health NHS Trust, London, UK
  2. 2London Ambulance Services, London, UK
  1. Correspondence to Dr S Honarbakhsh, St Bartholomew's Hospital, Barts Heart Centre, West Smithfield, London EC1A 7BE, UK; sherry0508{at}


Introduction Supraventricular tachycardias (SVTs) are a common cause of acute hospital presentations. Adenosine is an effective treatment. To date, no studies have directly compared paramedic-with hospital-delivered treatment of acute SVT with adenosine.

Method Randomised controlled trial comparing the treatment of SVT and discharge by paramedics with conventional emergency department (ED)-based care. Patients were excluded if they had structural heart disease or contraindication to adenosine. Discharge time, follow-up management, costs and patient satisfaction were compared.

Results Eighty-six patients were enrolled: 44 were randomised to paramedic-delivered adenosine (PARA) and 42 to conventional care (ED). Of the 37 patients in the PARA group given adenosine, the tachycardia was successfully terminated in 81%. There was a 98% correlation between the paramedics' ECG diagnosis and that of two electrophysiologists. No patients had any documented adverse events in either group. The discharge time was lower in the PARA group than in the ED group (125 min (range 55–9513) vs 222 min (range 72–26 153); p=0.01), and this treatment strategy was more cost-effective (£282 vs £423; p=0.01). The majority of patients preferred this management approach. Being treated and discharged by paramedics did not result in the patients being less likely to receive ongoing management of their arrhythmia and cardiology follow-up.

Conclusions Patients with SVT can effectively and safely be treated with adenosine delivered by trained paramedics. Implementation of paramedic-delivered acute SVT care has the potential to reduce healthcare costs without compromising patient care.

Trial registration number NCT02216240.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Adenosine is an effective treatment for supraventricular tachycardias (SVTs) excluding atrial fibrillation/flutter and atrial tachycardia, with high rates of conversion to sinus rhythm.1 ,2 Small prospective studies have demonstrated that administration of adenosine in a community setting is safe2 ,3 and there is a good correlation between paramedics' interpretation of the ECG and that of physicians.4 ,5 In cases where there has been inappropriate administration of adenosine, this had not been associated with adverse outcomes2 ,6 and, with training, inappropriate administration rates are reduced.7 To date, all studies examining the use of adenosine in the prehospital setting required all patients to be taken to hospital even if the arrhythmia had successfully been terminated. We hypothesised that treatment and discharge of a patient on location by a paramedic is safe and effective and is faster and more cost-effective than conventional emergency department (ED)-based care.


Patients presenting with regular narrow complex tachycardia on 12-lead ECG who were haemodynamically stable, with no history of structural or ischaemic heart disease and without contraindication to adenosine were considered for enrolment. The full inclusion and exclusion criteria are presented in box 1. Paramedics enrolled and randomised patients to paramedic treatment (PARA) or standard ED-based care. Because patients were likely to have distressing symptoms, they gave verbal consent before randomisation and then written consent after treatment. The study was conducted in accordance with the Declaration of Helsinki8 and was reported according to the CONSORT guidelines.9

Box 1

Inclusion and exclusion criteria for enrolment in the study

Inclusion criteria

▸ Age above 16 years old

▸ 12-lead ECG showing a regular narrow complex tachycardia

Exclusion criteria

▸ Age less than 16 years old

▸ Already enrolled in the study or another research study

▸ Inability or unwillingness to give consent

▸ Syncope or blackout

▸ Sinus tachycardia

▸ ECG evidence of myocardial infarction

▸ Broad complex tachycardia

▸ History of illicit drug use within the previous 24 hours

▸ History of structural or ischaemic heart disease

▸ Pregnancy

▸ Failure to obtain intravenous access

▸ Hypotension (systolic blood pressure of <100 mm Hg)

▸ Contraindication to adenosine (asthmatic on regular inhalers, use of dipyridamole and history of heart transplant)

Paramedics participating in the study had prior cardiac training: they were enrolled on a 5-hour course which included research governance training, a cardiac arrhythmia update, the pharmacology of adenosine, and a knowledge assessment at the end.

The PARA group was treated at the scene with Valsalva manoeuvres followed by administration of 6 and 12 mg adenosine unless the tachycardia terminated. Patients were taken to the ED if the tachycardia did not terminate, restarted, or the patient had continuing symptoms, a persistently abnormal ECG (other than T wave inversion) or was haemodynamically unstable. If this was not the case, the paramedic discharged the patient at the scene. After paramedic discharge, patients received an information pack and a referral letter for their family doctor to refer them to a cardiology clinic, along with a copy of their ECG. The patients in the ED group were treated as per standard care and transferred to their nearest ED.

Follow-up data were obtained at 4 weeks and 6 months via telephone conversation with the patient. At follow-up, patients were asked whether they felt satisfied with the treatment they received. Patients were also asked if they would have preferred receiving their treatment as per the alternative group.

The primary end point was the time from the first point of contact to discharge from medical care, whether this was at the scene by paramedics or in hospital. Secondary end points were correct ECG diagnosis, complication rates of treatment, whether an ECG was given to the patient at point of discharge, whether antiarrhythmic medications were taken and whether cardiology review had taken place by 6 months.

In addition, a cost analysis was performed comparing the cost of treatment for each group. For each group the ambulance call-out costs were calculated. For those who presented to the ED, the cost of presentation to the ED together with investigation costs was determined. Arrhythmia-related hospital admission costs were also determined.

Statistical analysis was performed using SPSS V.20. A p value of <0.05 was taken to indicate significance. Normally distributed data were analysed using a t-test, and data were presented as mean±SD. To determine whether data were normally distributed, we used both histogram plots that were compared with a ‘bell curve’ plot and the Shapiro-Wilk test. Non-normally distributed data were analysed using the Mann–Whitney U test, and data were presented as median and range. Proportional data were compared using Fisher's exact test. Patients randomised in contradiction to the inclusion/exclusion criteria and those where there was a protocol violation were included in an intention-to-treat analysis. The study was powered to assess differences in discharge times between the two groups. We hypothesised that the mean discharge time would be 2160±780 min and 1728±680 min in the ED and PARA group, respectively. By using two-sampled t–tests, an α of 0.05 and an 80% power, we found that a total of 90 patients (45 in each group) would be required to achieve statistical significance.

The study was approved by the local ethics board and was prospectively registered (Clinical identifier NCT02216240).

Patient involvement

Patients were not involved in the predesign of this study.


Between 5 December 2010 and 28 September 2014, 90 patients were screened. Four patients did not meet the criteria, and 86 patients were randomised (figure 1). Forty-two patients were randomised to the ED group and 44 to the PARA group (figure 1). Eight patients (9.3%) were inappropriately randomised to the study—that is, the inclusion and exclusion criteria were not met. There was a protocol violation in four patients (4.7%) (figure 2).

Figure 1

Flow diagram showing the enrolment and randomisation of patients included in the study and the number of patients who were either discharged by the emergency department (ED) or admitted as an inpatient.

Figure 2

Number of patients who were inappropriately randomised to the study or there was a protocol violation.

The baseline characteristics of the cohort are presented in table 1. A large proportion of the patients had a previous history of similar symptoms (70/86, 81%), with many of them having previously sought medical attention (66/86, 77%) and/or seen a cardiologist (45/86, 52%). Half of these patients had been in contact with paramedics previously (43/86, 50%), many of them more than once (21/86, 24%). There was no significant difference in these findings between the two groups.

Table 1

Baseline characteristics of the cohort and the management of the SVT

In the PARA group, adenosine was given to 37 of the 44 patients (84%; table 1). For the seven cases where adenosine was not administered, this was because of spontaneous termination (three), termination by vagal manoeuvres (one), underlying typical right atrial flutter revealed by vagal manoeuvres (one), inability to achieve intravenous access (one), and adenosine not available (one). Adenosine successfully terminated the tachycardia in 30 cases without any re-initiation (30/37, 81%), with 77% of these patients (23/30) being discharged directly by paramedics. Those who were taken to the ED (one for persistent sinus tachycardia after adenosine, five for ECG changes after conversion, and one for broad complex tachycardia) were all safely discharged. On review of the PARA group as a whole, 25 patients (57%) were discharged directly by the paramedics, and 19 were taken to the ED. All of the reasons for admission to the ED are shown in table 2 and figure 3. Of the 19 patients in the PARA group who were taken to the ED, 10 (53%) had additional treatment in the ED, while the remaining nine were discharged.

Table 2

Reasons for admission to the ED for the patients in the PARA group (N=44)

Figure 3

Reasons for admission to the emergency department (ED) in the group treated and discharged by paramedics.

The ED group was less likely to be given adenosine for the treatment of tachycardia (84% vs 60%; p=0.02; table 1). The main reason why adenosine was not provided was that the tachycardia spontaneously terminated either on/before arrival or with vagal manoeuvres (14/25, 56%; p=0.01). If these patients were excluded, there was no significant difference in adenosine administration (p=0.74). In the ED group, where tachycardia had not terminated spontaneously, Valsalva manoeuvres and adenosine successfully terminated the tachycardia in 27 of the remaining 30 patients (90%).

Primary end point

The median total time from first point of contact to discharge from medical care was 125 min (range 55–9513) in the PARA group and 222 min (range 72–26 153) in the ED group (p=0.01).

In the PARA group, the median time to discharge was much shorter in those who were discharged by paramedics at the scene (105 min, range 55–220) compared with those taken to hospital (297 min, range 154–9513) (p=3.1231556E-11).

Secondary end points

Safety of treatment and accuracy of the diagnosis

All patients had adenosine administered safely with no unexpected events.

The diagnosis made by the paramedics on reviewing the ECGs correlated with that made by two electrophysiologists in all except two cases: a broad complex tachycardia that was inappropriately randomised to the study and a diagnosis of atrial fibrillation (84/86, 98%).


Those in the PARA group were more likely to receive a copy of their ECG at the point of discharge than those in the ED group (34/40 vs 22/35; p=0.035) (table 3).

Table 3

Follow-up findings at 4 weeks and 6 months for the two groups

At 6 months follow-up, 35 patients in the PARA group had been seen by a cardiologist and 23 had been offered a catheter ablation. This was not significantly different from that seen in the ED group (table 3). Further to this, there was no significant difference between the two groups in the number of patients who had an echocardiogram or thyroid function tests (TFTs), with those in PARA group being more likely to have had TFTs checked (59% vs 29%, p=0.02).

Patient satisfaction

In the PARA group, 29 patients (29/44, 66%) felt satisfied and preferred having received adenosine from paramedics, and this included 20 of the 25 patients who were discharged by paramedics (80%). Only three patients would have preferred to have had their treatment in the ED, only one of whom was discharged by paramedics. These three patients felt the ED would be a safer environment. The remaining patients had no preference either way. In the ED group as many as 33% felt that having had their treatment delivered by the paramedics might have been beneficial. Twenty-four patients preferred to have received their treatment in the ED, while the remaining patients had not thoughts either way.

Cost analysis

The cost analysis included ambulance call-out costs, ED presentation costs, the costs of investigations performed and the cost of arrhythmia-related hospital admission. The median cost in the PARA group was £282 (range 282–555) and that of the ED group was £423 (range 361–854) (p=0.01).


This is the first study to compare paramedic treatment and discharge of regular narrow QRS complex SVT with standard care. With training, paramedics were able to appropriately diagnose SVT, select patients appropriately for therapy, and administer adenosine safely. Compared with conventional management, management by paramedics was associated with a significantly reduced time to discharge, and hospital attendance was completely avoided in 57%. This resulted in reduced cost and inconvenience for the patient and did not affect long-term access to treatment such as catheter ablation.

In the ED group, the majority of patients had their tachycardia terminated spontaneously, by vagal manoeuvres or through the use of adenosine, after which they were discharged from the ED (32/35, 91%). This further emphasises the number of patients that could be effectively treated by paramedics and not require presentation to a hospital. The paramedics transferred a large proportion of the patients to the ED after administration of adenosine. However, of these, only 52% received additional treatment there. This highlights that there is an important role for the ED in the management of these patients and an all-or-nothing approach with regard to paramedic treatment of SVTs should not be advocated.

Patients discharged by paramedics were significantly more likely to receive a copy of their ECG showing the arrhythmia (p=0.035). As the ongoing management of SVT is highly dependent on cardiologists having the ability to review the ECG, this clearly has an impact on the future management of these patients. We have demonstrated that being treated and discharged by paramedics did not result in the patients being less likely to receive ongoing management of their arrhythmia and cardiology follow-up (p=0.22). This suggests that the information provided by the paramedics was sufficient for their family doctor to use when referring these patients to secondary or tertiary care. There was also no significant difference between the groups in the proportion of patients having had echocardiograms, TFTs or being on antiarrhythmic or rate-controlling drugs at the 4-week follow-up. This suggests treatment and discharge by a paramedic does not delay or prejudice subsequent elective investigation.

A previous study demonstrated that the diagnosis made by paramedics agrees with that made by physicians.7 This was further demonstrated in our study, where 98% of the ECG diagnoses made by the paramedics agreed with those of two electrophysiologists. This confirms that paramedics are able to effectively and safely diagnose and treat regular, narrow complex SVTs.


This trial was necessarily performed on site by paramedics without research staff or clinicians present. It was therefore arguably expected that a number of protocol violations would occur. Nevertheless, this forms an important part of the dataset, as testing the feasibility of a paramedic-led approach was a fundamental part of the study. In general, it seems that this approach was followed safely and effectively.

All of the paramedics involved in this study underwent training before their involvement; therefore our findings apply to trained paramedics.

The nature of the study did not allow blinding of research staff or patients. It must also be acknowledged that the study was conducted in a large UK city (London) and aspects of the results may have differed if the study were conducted elsewhere. However, the results would arguably be similar for much of the Western world.


This study demonstrates that the management of patients with benign SVT can effectively and safely be delivered by trained paramedics. This approach was associated with reduced time to discharge and avoids the need for hospital admission at all in the majority of patients. Paramedic management was associated with reduced cost and greater patient satisfaction compared with conventional hospital-based management. Implementation of paramedic-delivered acute SVT management can reduce healthcare costs for these patients without adversely affecting their short- or long-term management.

Key messages

What is already known on this subject?

  • Adenosine is an effective treatment for supraventricular tachycardias. Small prospective studies have shown that administration of adenosine in a community setting is safe.

What might this study add?

  • We have shown that prehospital paramedic-delivered treatment is associated with significantly shorter discharge times compared with conventional emergency department-based care. This treatment strategy is also associated with greater patient satisfaction and is more cost-effective.

How might this impact on clinical practice?

  • Patients attending with benign tachycardia can often undergo unnecessary hospital stays and investigations, which may delay their definitive therapy by a specialist. We have demonstrated that paramedic-delivered treatment has the ability to overcome these issues without compromising patient safety.



  • Contributors SH analysed the data collected, conducted the statistical analysis, and wrote the first and final draft of the manuscript. VB and RJS formulated the research idea, ran the research study, and gave critical appraisal of the manuscript. RJH and MF were involved with the study design and review of the data as well as critically appraising the manuscript. CK, KP, WU and MJE participated in the data collection process and critical appraisal of the manuscript. GR, SA and LR took part in training the paramedics involved in this study.

  • Funding The study was funded by the British Heart Foundation.

  • Competing interests None declared.

  • Transparency declaration The corresponding author affirms that the manuscript is an honest, accurate, and transparent account of the study being reported; no important aspects of the study have been omitted, and any discrepancies from the study as planned have been explained.

  • Ethics approval Local ethics board at Barts Health.

  • Provenance and peer review Not commissioned; externally peer reviewed.