Author + information
- Received February 21, 2007
- Revision received April 26, 2007
- Accepted April 30, 2007
- Published online August 7, 2007.
- Vivek N. Dhruva, DO⁎,
- Samir I. Abdelhadi, DO⁎,
- Ather Anis, MD⁎,
- William Gluckman, DO, FAEP†,
- David Hom, MS‡,
- William Dougan, MICP†,
- Edo Kaluski, MD, FACC⁎,
- Bunyad Haider, MD, FACC⁎ and
- Marc Klapholz, MD, FACC⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Marc Klapholz, UMDNJ—New Jersey Medical School, 185 South Orange Avenue, MSB I-538, Newark, New Jersey 07103.
Objectives Our goal was to examine the effects of implementing a fully automated wireless network to reduce door-to-intervention times (D2I) in ST-segment elevation myocardial infarction (STEMI).
Background Wireless technologies used to transmit prehospital electrocardiograms (ECGs) have helped to decrease D2I times but have unrealized potential.
Methods A fully automated wireless network that facilitates simultaneous 12-lead ECG transmission from emergency medical services (EMS) personnel in the field to the emergency department (ED) and offsite cardiologists via smartphones was developed. The system is composed of preconfigured Bluetooth devices, preprogrammed receiving/transmitting stations, dedicated e-mail servers, and smartphones. The network facilitates direct communication between offsite cardiologists and EMS personnel, allowing for patient triage directly to the cardiac catheterization laboratory from the field. Demographic, laboratory, and time interval data were prospectively collected and compared with calendar year 2005 data.
Results From June to December 2006, 80 ECGs with suspected STEMI were transmitted via the network. Twenty patients with ECGs consistent with STEMI were triaged to the catheterization laboratory. Improvement was seen in mean door-to-cardiologist notification (−14.6 vs. 61.4 min, p < 0.001), door-to-arterial access (47.6 vs. 108.1 min, p < 0.001), time-to-first angiographic injection (52.8 vs. 119.2 min, p < 0.001), and D2I times (80.1 vs. 145.6 min, p < 0.001) compared with 2005 data.
Conclusions A fully automated wireless network that transmits ECGs simultaneously to the ED and offsite cardiologists for the early evaluation and triage of patients with suspected STEMI can decrease D2I times to <90 min and has the potential to be broadly applied in clinical practice.
The benefit of decreasing door-to-balloon times (D2B) in acute ST-segment elevation myocardial infarction (STEMI) has previously been widely reported (1–4). Strategies to decrease D2B times are often limited by multiple time-intensive steps. Implementation of wireless electrocardiogram (ECG) transmissions from the field to the emergency department (ED) have recently been shown to significantly decrease D2B times but still require personnel in the ED to forward the ECG to the cardiologist (5). The STAT-MI project (ST-Segment Analysis Using Wireless Technology in Acute Myocardial Infarction) was devised to demonstrate that newer wireless technologies and software linked in a network can facilitate simultaneous transmission of high-resolution ECGs to both the ED and offsite cardiologists and further improve door-to-intervention (D2I) times. We report on the initial results of implementing this completely automatic notification and response system for the evaluation and management of STEMI and its use in reducing D2I times.
The STAT-MI network was devised by a collaborative task force at the University of Medicine and Dentistry, New Jersey—University Hospital led by cardiology staff and including members of emergency medical services (EMS), ED, hospital administration, hospital information and technology, telecommunications, medical informatics, and physician staff. Representatives from Medtronic Corporation (LIFEPAK 12, LifeNet Receiving Station, Minneapolis, Minnesota) and Verizon Wireless (phones, wireless network, paging) assisted in the implementation of this network.
From June to December 2006, all patients identified by EMS personnel in the field with suspected STEMI were included using the newly designed STAT-MI network. All patients confirmed to have STEMI were brought to the cardiac catheterization (cath) lab for primary percutaneous intervention (PCI). All STEMI patients transported by EMS in the previous year were used as historical controls.
This fully automated network, initiated through the acquisition of an ECG by EMS personnel in the field, is shown in Figure 1.Bluetooth-enabled LIFEPAK 12 defibrillators (Medtronic) automatically transmitted the 12-lead ECG to Bluetooth-enabled Motorola E815 phones (Verizon Wireless) worn by EMS personnel. E815 phones automatically transmitted the 12-lead ECGs to the LifeNet Receiving Station (Medtronic), which automatically printed the ECG for local review, converted the ECG to Adobe PDF format, sent the ECG via University Hospital’s secure intranet to prespecified e-mail addresses of the cardiologists, and sent an audible text page to the on-call cardiologist. No intervening action by any personnel was required once EMS activated the system in the field. On-call cardiologists accessed via their handheld smartphones (Audiovox XV6700, Verizon Wireless) the e-mail with the attached PDF ECG file. The e-mail communication program used was NotifyLink (Notify Technology Corporation, San Jose, California). Electrocardiograms were transmitted with an embedded telephone number for the en route EMS personnel allowing for direct communication before hospital arrival (Fig. 2).The ECG could be magnified and moved on the smartphone screen allowing for detailed ST-segment and other waveform analyses.
Based on the ECG and en route discussion with EMS, a decision for primary PCI was made. During regular hours (7 amto 5 pm, Monday to Friday), the patient was brought directly to the cardiac cath lab. During off hours (5 pmto 7 amand weekends), the cath lab team would be notified, and the patient would be delivered to the ED if the cath lab team had not yet arrived.
Demographic and clinical data was collected on all patients who were triaged to the cath lab and underwent primary PCI. Time data were collected beginning at the time that the ECG was transmitted. Door-to-cardiologist-notification timewas the time from hospital arrival to the notification of the cardiologist on call. If notification occurred before hospital arrival as expected with this network, a negative value was assigned. Door-to-arterial access (D2A) timewas the time between hospital arrival and femoral arterial access. Time-to-first angiographic injectionwas the time from hospital arrival to the first intracoronary injection of contrast. Time-to-wire across lesionwas the time between hospital arrival and the passage of the guidewire across the culprit lesion. Door-to-intervention timewas the time between hospital arrival and the first intervention that restored patency of the culprit vessel. The term “intervention” as opposed to “balloon” was chosen as more congruent with “primary percutaneous intervention” and could include angioplasty, stenting, and/or mechanical thrombectomy. Patients with suspected STEMI but without an interventional procedure were included in the analysis of D2A times.
Primary end point
The primary outcome was to compare D2I times for primary PCI using the STAT-MI network as compared with D2I times in calendar year 2005.
Secondary end points
Changes in the other time intervals noted in the preceding text.
Analyses included descriptive statistics, ttest for independent means, chi-square, and Fisher exact test for categorical analysis and Wilcoxon rank sum nonparametric analysis for comparing door-to-cardiologist notification time, which had a nonparametric range of time intervals from −26 to 280 min. Other measurements of time interval are presented as means with standard deviations in minutes. Analysis of variance using treatment groups (STAT-MI or 2005 control subjects) and time of presentation (day or night/weekends) as fixed factors examined independent and interaction effects when looking at key time measures. Statistical analysis was conducted using SPSS/PC+ V15.0 for Windows (SPSS Inc., Chicago, Illinois). Statistical significance was set at a value of p < 0.05.
Between June and December 2006, 80 ECGs for suspected STEMI were transmitted from EMS personnel to the cardiologist on call via the STAT-MI network. Of these, 20 (25%) were interpreted as consistent with STEMI, 11 (14%) nonspecific ST-segment or T-wave abnormalities, 11 (14%) supraventricular tachycardia, 10 (12%) age-indeterminant Q-wave myocardial infarction, 8 (10%) right bundle branch block/intraventricular conduction defect, 6 (7%) left ventricular hypertrophy, 3 (4%) ST-segment depressions suggestive of acute coronary syndrome, 3 (4%) sinus bradycardia, 1 (1%) paced rhythm, and 7 (9%) normal. Median time from ECG acquisition in the field to remote desktop availability was 2 min and to visualization on smartphones 4 min. None of the 60 patients with non-STEMI ECGs sent to the ED was subsequently found to have STEMI. No STEMIs were missed because of failure to transmit from the field. One STEMI patient was excluded because of a cardiac arrest and prolonged resuscitation before intervention. Of the patients brought to the cardiac cath lab, 3 had angiographically normal coronary arteries, and 1 had nonobstructive coronary artery disease.
Baseline characteristics for the 20 patients and 29 historical controls are shown in Table 1.There were no significant differences between the 2 groups. The majority of patients were men, and the most common antecedent cardiac risk factors were hypertension and cigarette smoking.
Key time measure interval data
There was significant improvement in all key time interval measures (Table 2).Mean door-to-cardiologist notification time was reduced by 76 min (61.4 to –14.6 min, p < 0.001), and mean D2A time was reduced by 60 min (108.1 to 47.6 min, p < 0.001). The D2I time was reduced by 66 min (145.6 to 80.1 min, p < 0.001).
Time interval measured by arrival time
Twelve (60%) patients arrived during regular hours in the STAT-MI group compared with 14 (48%) patients in the control group. Eight (40%) patients arrived during off hours in the STAT-MI group compared with 15 (52%) patients in the control group. Regardless of the time of day that patients presented, the magnitude of benefit of the STAT-MI network in reducing D2I times was similar (Table 3).No interaction between group assignment and time of presentation was observed in any of the key time measures.
Shortening of D2B times remains a singular focus in the management of STEMI (6), with current American College of Cardiology/American Heart Association (ACC/AHA) guidelines recommending a D2B time of <90 min for mortality benefit (7). Using the STAT-MI network, significant improvement was seen in all measured time variables with mean D2I times of 80.1 min.
The preponderance of improvement in D2I time in our study occurred as a consequence of much earlier notification of the cardiologist on call—on average 15 min before patient arrival to the hospital. Time delays in patient evaluations and ECG acquisition and transmissions in active EDs can delay the timely management of STEMI patients (8). Wireless transmission of prehospital ECGs to the ED has helped decrease D2B times and is now widely recommended (9). In the National Registry of Myocardial Infarction (NRMI), a prehospital ECG is associated with improved D2B times (94.0 vs. 110.3 min) (10). Our study took further advantage of wireless technology by completely automating the process of ECG acquisition and transmission from the field to a handheld smartphone worn by cardiologists. Embedding the EMS phone number on the ECG was a further refinement that allowed for direct communication between the cardiologist and en route EMS personnel.
The reduction in D2I times was similar regardless of time of presentation. In a recent report from the NRMI registry evaluating time of day to D2B times, the shortest mean D2B time in any subgroup in the registry for regular and off hours was 86 and 108 min, respectively, as compared with our study of 63 and 103 min, respectively (11).
In November 2006, the ACC launched the GAP-D2B initiative, a national quality improvement campaign focused on shortening D2B times to 90 min or less. The D2B workgroup advocates 6 key strategies including ED physician activation of the cath lab, single-call activation of the cath lab, cath lab team arrival within 20 min, prompt data feedback to cath lab and ED, senior management commitment, and a team-based approach (12). Our study demonstrates that the application of wireless technology with automatic notification of cardiology from the field and direct routing of the patient to the cath lab produced significant time savings that can be additive to the strategies articulated by the D2B workgroup.
Comparison with historical controls can be flawed by selection bias and differences in interventional techniques or operator expertise. Our study and control populations were temporally contiguous (2006 vs. 2005), and review of our demographic data suggests that the patient populations were clinically comparable. All patients with STEMI in the year 2005 and transported via EMS were included in this study, and all study patients were consecutive. Furthermore, there were no apparent differences in cath techniques, as procedural success, patency rates, and case durations were similar between groups. The 61 min that it took to notify the on-call cardiologist in 2005 (control group) clearly exceeded recommendations from ACC/AHA practice guidelines.
A fully automated wireless network that transmits ECGs simultaneously to the ED and offsite cardiologists for the early evaluation, triage, and treatment of patients with STEMI can decrease D2I times to <90 min and has the potential to be broadly applied in clinical practice. This technology has moved the proverbial “door” to the patient’s home.
The authors would like to thank the following people at the University of Medicine and Dentistry, New Jersey—University Hospital for their assistance in developing and maintaining the STAT-MI network: Keith McCabe, Gary Lomonaco, Richard Tunnell, David Langley, Dr. Hosseinali Shahidi, Nancy Hamstra, Dr. Suzanne Atkin, Oscar West-Crews, Deborah Dolny-Korasick, EMS personnel, cardiac cath lab staff, cardiology fellows, and attending staff.
Grants for this study were received from the Verizon Foundation (Basking Ridge, New Jersey) and Medtronic Corporation (Minneapolis, Minnesota).
- Abbreviations and Acronyms
- American College of Cardiology/American Heart Association
- door-to-arterial access
- emergency department
- emergency medical services
- National Registry of Myocardial Infarction
- percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- Received February 21, 2007.
- Revision received April 26, 2007.
- Accepted April 30, 2007.
- American College of Cardiology Foundation
- De Luca G.,
- Suryapranata H.,
- Zijlstra F.,
- et al.,
- ZWOLLE Myocardial Infarction Study Group
- Berger P.B.,
- Ellis S.G.,
- Holmes D.R. Jr..,
- et al.
- McNamara R.L.,
- Wang Y.,
- Herrin J.,
- et al.
- Krumholz H.M.,
- Anderson J.L.,
- Brooks N.H.,
- et al.
- Antman E.M.,
- Anbe D.T.,
- Armstrong P.W.,
- et al.
- Pinto D.S.,
- Kirtane A.J.,
- Nallamothu B.K.,
- et al.
- Garvey J.L.,
- MacLeod B.A.,
- Sopko G.,
- et al.
- Curtis J.P.,
- Portnay E.L.,
- Wang Y.