doi: 10.1111/j.1742-6723.2007.00966.x Emergency Medicine Australasia (2007) 19, 229–233

© 2007 The AuthorsJournal compilation © 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

Blackwell Publishing AsiaMelbourne, AustraliaEMMEmergency Medicine Australasia1742-6731© 2007 The Authors; Journal compilation © 2007 Australasian College for Emergency Medicine and Australasian Society forEmergency Medicine200719••229233Original ArticlesTelemetry monitoring during transportA Lin

et al

.

Correspondence: Ms Debra Kerr, Joseph Epstein Centre for Emergency Medicine Research, 1st Floor, Sunshine Hospital, 176 Furlong Road, St Albans, Vic. 3021, Australia. Email: Debbie.Kerr@wh.org.au

Andrew Lin, Advanced Medical Science Student; Debra Kerr, BN MBL, Deputy Director; Anne-Maree Kelly, MD, FACEM, Director.

ORIGINAL RESEARH

Is cardiac monitoring during transport of low-risk chest pain patients from the emergency department necessary?Andrew Lin,1 Debra Kerr2 and Anne-Maree Kelly2

1The University of Melbourne, Melbourne and 2Joseph Epstein Centre for Emergency Medicine Research, Western Health, St Albans, Victoria, Australia

Abstract

Objective: Incidence of life threatening arrhythmia for patients who present to the ED with low-riskchest pain (CP) (non-ischaemic electrocardiograms and normal cardiac marker profiles) israre. These patients are often transported with cardiac monitoring by nurse escort fromthe ED. We aimed to show that this group of patients are at low risk of experiencing life-threatening arrhythmia disturbances.

Methods: This was a prospective, observational study of ED low-risk CP patients who presented inthe period September 2005 and March 2006 and were transported with cardiac monitoring.Data were collected via chart review, and nurse escorts prospectively documented trans-port details. The primary study outcome was the development of a life threateningarrhythmia requiring treatment during transport from the ED. Data analysis includeddescriptive statistics and interrater agreement.

Results: During the study period there were 231 patients admitted to monitored beds from the ED,of whom 170 (74%) were low risk and enrolled in the study. No patient sustained anadverse event during transport from the ED (0%; 95% confidence interval 0–2.2%). Mean(±SD) time required for nurse escort to and from the radiology department and coronarycare unit was 9.0 (±3.1) and 16 (±6.5) min, respectively.

Conclusions: CP patients who present to the ED with normal electrocardiograms and cardiac markerprofiles are at low risk (<1%) of experiencing an adverse event during transport from theED. This subset of patients might not require cardiac monitoring or nurse escort duringtransportation from the ED.

Key words: arrhythmia, chest pain, electrocardiography, telemetry.

A Lin et al.

230 © 2007 The AuthorsJournal compilation © 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

Introduction

A liberal practice of hospital admission and cardiacmonitoring for chest pain (CP) patients who present tothe ED persists, presumably for fear of consequencesfor the patient (morbidity/mortality) and physician(legal). The risk-adjusted mortality for patients withmissed MI diagnoses who are inadvertently dis-charged is almost double that of patients who arediagnosed and hospitalized.1 Professional guidelinesrecommend a period of assessment with continuouscardiac monitoring for patients who are admitted withCP syndromes.2

There is an increasing amount of evidence to sug-gest that CP patients at low-to-intermediate risk ofdeveloping a life-threatening complication can besafely managed after hospital admission without con-tinuous cardiac monitoring after an initial observationperiod in the ED.3–7 Generally, these patients are sta-ble, with no ongoing CP or cardiac rhythm distur-bances, and have non-ischaemic electrocardiograms(ECG) and normal cardiac markers during their EDstay.

Cardiac monitoring is usually continued duringtransport of patients from the ED, to the radiologydepartment for a chest X-ray or to a monitored bed inthe coronary care unit (CCU). In addition, patients areoften accompanied by a nurse with skills in identifyingand managing life-threatening arrhythmias (AdvancedCardiac Life Support [ACLS]). This current protocol fortransporting patients is problematic, as delays in inves-tigations and transfers to CCU can occur while waitingfor availability of a portable cardiac monitor and amutually suitable time for transport between non-medical personnel (orderly) and an ACLS trained nurse.Moreover, in having to escort patients, nurses arediverted away from the ED and are unable to attend toother clinical responsibilities.

In two North American studies it has been shownthat transportation of low-risk CP patients from the EDwithout cardiac monitoring or nurse escort is safe, andhas the potential to reduce diversion of nurses fromthe ED.8,9

The present study aims to test the hypothesis thata subset of CP patients who present to the ED, whohave non-ischaemic ECG and normal cardiac mark-ers, are at low risk (<1%) of experiencing a life-threatening arrhythmia requiring treatment, andhence do not benefit from cardiac monitoring ornurse escort during transport to the radiology depart-ment or CCU.

Methods

This was a prospective, observational cohort studyundertaken in the ED of Western Hospital (WH), Foot-scray, Victoria, Australia, in the period September 2005to March 2006. WH is a 323-bed adult teaching hospitalwith an ED that sees approximately 34 000 patients ayear. Formal ethics approval was obtained from Mel-bourne Health Research Directorate, and individualpatient consent was not required.

This was a convenience sample. The majority ofcases were identified when the primary researcher (AL)was present in the ED, during Monday to Friday, from9.00 hours to 17.00 hours. Nurses were asked to identifyeligible subjects who presented after-hours and onweekends; times during which researchers were absentfrom the ED.

Patients with the chief presenting complaint of CP,epigastric pain, unstable angina, and angina wereassessed for study inclusion as identified from the EDinformation management system. Patients who had aCP complaint, received cardiac monitoring while in theED and during transport, and had low-risk criteria wereconsidered eligible for study inclusion. Low-risk criteriaincluded stable cardiac rhythm, non-ischaemic ECG (noST elevation or depression [>0.5 mm] or deep T waveinversion in more than two leads) and normal cardiacmarkers (creatine kinase [CK] <200 IU/L in men or<160 IU/L in women, cardiac troponin I <0.4 µg/L).Patients were considered ineligible for enrolment if theymet one or more of the following criteria during theirED stay: unstable cardiac rhythm disturbance (asystole,ventricular tachycardia or fibrillation, rapid atrial fibril-lation, 2nd or 3rd degree AV block), ischaemic ECG andelevated cardiac markers (as defined previously).

Data on all subjects were collected via chart andpathology database review, by one researcher (AL).Structured data extraction was performed using anexplicit data form for defined variables. For each case,a copy of the patient’s ECG was kept for interraterreliability assessment. Nurses escorts were asked torecord information about the occurrence and treatmentof any life-threatening arrhythmias or minor adverseevents during transport, along with the duration ofnurse escort.

Data abstracted from the patient history and pathol-ogy databases included patient demographics, past car-diac history, cardiac risk factors, symptoms, clinicalcondition during ED stay, risk stratification (accordingto the National Heart Foundation criteria2), laboratoryfindings (CK, troponin I) and ECG assessment.

Telemetry monitoring during transport

© 2007 The Authors 231Journal compilation © 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

The primary outcome measure was the developmentof a life-threatening arrhythmia requiring treatment dur-ing transport from the ED. A life-threatening arrhythmiawas defined as ventricular tachycardia, ventricularfibrillation, asystole, or tachy/bradycardia with cardiaccompromise. Treatment was defined as the administra-tion of medication, cardioversion, or cardiopulmonaryresuscitation (CPR). Also noted was the development ofany minor adverse event (CP, shortness of breath, syn-cope, hypotension) during transport from the ED.

Data were analysed using SPSS10 for descriptive sta-tistics (proportions, means, medians, 95% confidenceinterval [CI], SD). There was excellent agreement(kappa = 1.0) for ECG interpretation by two reviewers(AL, DK). No formal sample size calculation wasconducted.

Results

During the study period, we identified 231 CP patientswho were admitted to monitored beds in the ED and

transported with cardiac monitoring and nurse escort.Of these, 61 cases were ineligible for having one or moreof the exclusion criteria during their ED stay. The finalsample consisted of 170 patients, as shown in Figure 1.

The average age was 67 years; and there were moremen (66%) than women (34%) in the sample. Table 1describes demographic data, cardiac risk factor profiles,initial cardiac rhythm, CP management and disposition.There were 64 patients (38%) who were dischargedhome from the ED, but each received continuous cardiacmonitoring while in the ED and during transport fromthe ED to the radiology department.

Hypertension was the most common risk factor forCAD (61%). Seventy-seven patients (45%) had hyperc-holesterolaemia, and almost a third of patients had dia-betes mellitus (32%). Most patients had two or more riskfactors for CAD (60%). Almost half of all patients expe-rienced recurrent CP while in the ED (46%) and most(82%) were in normal sinus rhythm at ED presentation.

The mean (±SD) total time patients were on cardiacmonitoring during their ED stay was 7.1 (±4.5) h.This included time during transport to and from the

Figure 1. Sample.

11Ischemic ECG changes

2Unstable cardiac rhythm disturbance

andElevated cardiac markers

27Elevated cardiac markers

13Ischemic ECG changes

andElevated cardiac markers

6Unstable cardiac rhythm disturbance

2Ischemic ECG changes

andUnstable cardiac rhythm disturbance

170Patients met the inclusion criteria

-Normal ECG-Normal cardiac markers

231Patients admitted to monitored beds in the ED

and transported with cardiac monitoring

A Lin et al.

232 © 2007 The AuthorsJournal compilation © 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

radiology department; but did not include the period ofmonitoring during transfer to the CCU.

Of 170 patients, 163 (96%) were transported withcardiac monitoring and a nurse escort to the radiologydepartment for a chest X-ray and 70 (41.2%) were trans-ported to the CCU. No patient sustained a life-threateningarrhythmia requiring treatment during transport (95%CI 0–2.2%). Also, there were no minor adverse events.

The mean time required for an ACLS trained nurseto accompany a patient to and from the radiologydepartment for a chest X-ray was 9.0 (±3.1) min. Aver-age time taken for a nurse to escort a patient to the CCUand then return to the ED was 16 (±6.5) min.

Discussion

ECG monitoring for patients who present to the EDwith CP is often maintained throughout ED assessmentand management, during transport, and after admission

to a monitored bed in the CCU. Our study found a 0%(95% CI 0–2.2%) rate of life-threatening arrhythmias in170 patients with non-ischemic ECG and normal cardiacmarkers who were transported from the ED with car-diac monitoring.

Two recent North American studies have also inves-tigated the utility of cardiac monitoring during trans-port of low-risk CP patients from the ED.8,9 Singer et al.8

followed 322 patients transported without cardiacmonitoring or nurse escort, all of whom had normal/non-diagnostic ECG and normal cardiac markers. Nolife-threatening arrhythmias occurred during transport(0%; 95% CI 0–0.93%). Pines et al.9 observed 281patients transported with monitoring and nurse escort,of whom 60% had normal/non-diagnostic ECG and90% had normal cardiac markers. Likewise there wereno new life-threatening arrhythmias requiring treat-ment during transport (0%; 95% CI 0–1.1%).

A low threshold for hospital admission in patientswho present with suspected acute coronary syndromeshas been questioned lately for patients considered at lowrisk of experiencing a life-threatening event. In a studyof 1029 CP patients at low risk (classified by Goldmanrisk <8%, and normal cardiac markers) by Hollanderet al.,11 no patient experienced a ventricular arrhythmiarequiring treatment while receiving cardiac monitoring.Locally, this has been further supported by research inthis field by Kelly et al.,5 who found that there was a lowrate of death (0.3%, 1/319) for CP patients who presentedto the ED with normal cardiac markers and ECG.

National Heart Foundation (Australia) guidelines2

recommend continuous cardiac monitoring for CPpatients with ‘intermediate’ risk features, that includeage over 65 years (54%) and two or more risk factorsfor coronary artery disease (60%). Our study, however,demonstrated that patients with non-ischemic ECGand normal cardiac markers are at low risk for life-threatening arrhythmias during transport from the ED,and might not require cardiac monitoring during thisphase. A significant number of patients (38%) weredischarged without hospitalization, suggesting thatthey might not have required continuous cardiacmonitoring during their ED stay.

The radiology department of WH ED is located adja-cent to the ED. Given its proximity, and the brief escorttime (9 min as evidenced in this present study), it mightbe safe to transport patients with non-ischemic ECG tothe radiology department without monitoring or nurseescort after retrieving normal cardiac marker results.

Recommendations regarding cardiac monitoring forpatients transported to the CCU of the study institution

Table 1. Patient characteristics

Age (years) mean ± SD 67 ± 15.3Sex

Male 112 (65.9%)Female 58 (34.1%)

Cardiac risk factorsFamily history of CAD 31 (18.2%)Hypertension 104 (61.2%)Hypercholesterolaemia 77 (45.3%)Obesity 15 (8.8%)Diabetes 54 (31.8%)Smoking 19 (11.2%)

Number of risk factors0 25 (14.7%)1 43 (25.3%)≥2 102 (60.0%)

RhythmNormal sinus rhythm 139 (82.1%)Sinus tachycardia 10 (5.9%)Sinus bradycardia 7 (4.1%)Atrial fibrillation 14 (8.2%)

Chest pain managementNo treatment required 96 (56.5%)Anginine 41 (24.1%)Anginine and morphine 32 (18.8%)I.v. glyceryl trinitrate ± heparin 11 (6.5%)

DispositionAdmission to CCU 70 (41.2%)Admission to another ward 27 (15.9%)Transfer to another hospital 9 (5.3%)Discharge home 64 (37.6%)

Telemetry monitoring during transport

© 2007 The Authors 233Journal compilation © 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

are more complex. The CCU is situated on the secondfloor of WH, and is only accessible via a lift locatedoutside the ED. Thus, there is a risk, albeit a very smallone, of lift failure and of the patient being trappedduring transport. If a life-threatening arrhythmia wereto occur in this instance, the patient would be in anenclosed space, inaccessible by doctors and nurses out-side. The decision to apply or withhold cardiac moni-toring during transport should be made by theattending physician.

There are benefits of a policy that does not demandcardiac monitoring for low-risk CP patients duringtransfer from the ED. Experienced nurses are notdiverted away from the ED, and transfer to inpatientbeds can be expedited because orderlies do not have towait for a nurse escort. Transport with an automaticexternal defibrillator, without cardiac monitoring, canalso be considered.

The present study is limited by the small sample size.A larger sample of at least 470 patients would berequired to reduce the upper 95% CI limit to below 1%for evidence of serious adverse event occurrence. Thiswas a convenience sample, in that not all eligiblepatients who presented to the ED during the studyperiod were enrolled. Data collection was limited to thetimes during which one of the researchers (AL) waspresent in the ED. However, this is unlikely to haveintroduced any selection bias into our sample. Ourstudy might have also suffered from observer bias. Incalculating escort times, researchers relied on nursesrecording the times they left and returned to the EDwhen they transported patients. We were unable toassess the accuracy of data in these situations. Finally,our study was observational, and we were only able tofollow the course of patients transported from the EDwith cardiac monitoring and nurse escort. Current hos-pital policy would not allow an experimental protocolof transporting patients without monitoring to betested. Ideally, a randomized control trial should beconducted, comparing actual outcomes of patientstransported with cardiac monitoring and nurse escortversus those transported without cardiac monitoring.

Conclusion

The present study found that patients who present tothe ED with CP, who have non-ischemic ECG and normalcardiac markers, are at low risk (<1%) of experiencinga life-threatening arrhythmia requiring treatment duringtransport from the ED. This subset of patients might

not require cardiac monitoring or nurse escort duringtransportation to the radiology department or CCU.

Acknowledgements

Thanks are extended to the nursing staff of the WesternHospital Emergency Department who identified andenrolled patients into the study.

Competing interests

None declared.

Accepted 25 January 2007

References

1. Pope JH, Aufderheide TP, Ruthazer R et al. Missed diagnoses ofacute cardiac ischemia in the emergency department. N. Engl. J.Med. 2000; 342: 1163–70.

2. Aroney C, Boyden AN, Jelinek MV et al. Management of Unsta-ble Angina Guidelines – 2000. National Heart Foundation ofAustralia and the Cardiac Society of Australia and New Zealand.Med. J. Aust. 2000; 173 (Suppl.): S65–88.

3. Hollander JE, Valentine SM, McCuskey CF, Brogan GX Jr. Aremonitored telemetry beds necessary for patients with non-traumatic chest pain and normal or non-specific electrocardio-grams? Am. J. Cardiol. 1997; 79: 1110–1.

4. Snider A, Papaleo M, Beldner S et al. Is telemetry monitoringnecessary in low-risk suspected acute chest pain syndromes?Chest 2002; 122: 517–23.

5. Kelly AM, Kerr D. Is it safe to manage selected patients withacute coronary syndromes in unmonitored beds? J. Emerg. Med.2001; 21: 227–33.

6. Kelly AM, Kerr D. Clinical features in the ED can identifypatients with suspected acute coronary syndromes who are safefor care in unmonitored hospital beds. Intern. Med. J. 2004; 34:594–7.

7. Sultana RV, Kerr D, Kelly AM, Cameron P. Validation of a toolto safely triage selected patients with chest pain to unmonitoredbeds. Emerg. Med. (Fremantle) 2002; 14: 393–9.

8. Singer AJ, Visram F, Shembekar A, Khwaja M, Viccellio A.Telemetry monitoring during transport of low-risk chest painpatients from the emergency department: is it necessary? Acad.Emerg. Med. 2005; 12: 965–9.

9. Pines JM, Rich VL, Datner EM et al. Utility of telemetry andnursing presence during transfer of admitted ‘rule out’ acutecoronary syndrome patients to inpatient beds. Acad. Emerg.Med. 2005; 12 (Suppl. 1): Abstract 208.

10. SPSS. spss for Windows, [computer software], Version 11.1.1.Chicago: SPSS, 2001.

11. Hollander JE, Sites FD, Pollack CV, Shofer FS. Lack of utility oftelemetry monitoring for identification of cardiac death and life-threatening ventricular dysrhythmias in low-risk patients withchest pain. Ann. Emerg. Med. 2004; 43: 71–6.