Author + information
- Received November 13, 2009
- Revision received April 15, 2010
- Accepted April 18, 2010
- Published online July 1, 2010.
- Hendrik-Jan Dieker, MD⁎ (, )
- Stephan S.B. Liem, MD,
- Hamza El Aidi, MD,
- Pierre van Grunsven, MD,
- Wim R.M. Aengevaeren, MD, PhD,
- Marc A. Brouwer, MD, PhD and
- Freek W.A. Verheugt, MD, PhD
- ↵⁎Reprint requests and correspondence:
Dr. Hendrik-Jan Dieker, Radboud University Nijmegen Medical Center, Heart Lung Center, Department of Cardiology 670, Experimental Cardiology, PO Box 9101, 6500 HB Nijmegen, the Netherlands
Objectives We sought to study the impact of direct referral to an intervention center after pre-hospital diagnosis of ST-segment elevation myocardial infarction (STEMI) on treatment intervals and outcome.
Background Primary angioplasty has become the preferred reperfusion strategy in STEMI. Ambulance diagnosis and direct referral to an intervention center is an attractive treatment option that has not been studied extensively.
Methods Consecutive pre-hospital patients with STEMI, who were referred to our intervention center for primary angioplasty between 2005 and 2007, were studied. After pre-hospital diagnosis, patients were either directly transported to our center or referred through a nonintervention center. The catheterization laboratory was activated before transport to the intervention center.
Results Of the 581 patients referred, 454 (78%) came with direct transport and 127 (22%) through a nonintervention center. Direct transport was associated with a higher proportion of patients treated within the 90-min time window of the STEMI guidelines: 82% versus 23% (p < 0.01). Patients directly transported had a significantly shorter median symptom-to-balloon time of 149 min (Interquartile range: 118 to 197 min) versus 219 min (interquartile range: 178 to 315 min), p < 0.01, a higher post-procedural Thrombolysis In Myocardial Infarction (TIMI) flow grade 3 rate (92% vs. 84%; p = 0.03), and a lower 1-year mortality rate (7% vs. 13%; p = 0.03). Direct transport to the intervention center was independently associated with the symptom-to-balloon time, which in turn was an independent predictor of post-procedural TIMI flow grade 3, a strong prognosticator of outcome.
Conclusions After ambulance-based diagnosis of STEMI, direct transport to an intervention center with pre-hospital notification of the catheterization laboratory more than triples the proportion of patients treated within the time window of the guidelines. Time to balloon was an independent predictor of post-procedural TIMI flow grade 3, which underscores the need to reduce treatment delays.
Primary angioplasty is the preferred reperfusion therapy in ST-segment elevation myocardial infarction (STEMI) if performed within 90 min of first medical contact by an experienced team of personnel in a high-volume center (1). In real-world practice, only ∼10% of patients in the U.S. meet the current time goal in case of referral for primary angioplasty (2). Strategies to reduce treatment delays in primary angioplasty are a subject of interest. It was shown that early activation of the catheterization laboratory is associated with a reduced door-to-balloon time (3). Pre-hospital diagnosis, notification, and direct referral to an intervention center is a promising strategy associated with a significant reduction in treatment delays (4,5). This strategy was associated with better left ventricular function and a lower risk of death or myocardial infarction (6) as compared to a strategy of referral from a nonintervention center. Although most studies using pre-hospital triage were performed in Europe, data from the U.S. confirm the feasibility and impact of the pre-hospital electrocardiogram in managing patients with STEMI (7–9). In the Netherlands, where distances between hospitals are much smaller and patients with suspected STEMI usually bypass the emergency physician and are immediately triaged by a cardiologist, a higher proportion of patients is treated within the time goal of the guideline. In the Nijmegen area (the Netherlands), a system of pre-hospital triage of STEMI has been used for over 15 years; it was initially designed to triage for pre-hospital fibrinolysis (10). This same system now serves for the triage for primary coronary intervention. In contrast to previous studies in which pre-hospital triage was compared with in-hospital triage, the current study focuses on pre-hospital triage and compares direct transport to an intervention center to referral through a nonintervention center. The aim of our study is to further elucidate the impact of direct transport to the intervention center on treatment intervals and procedural and clinical outcomes.
Consecutive patients with a pre-hospital diagnosis of STEMI referred for primary angioplasty to the catheterization laboratory of the Radboud University Medical Center in Nijmegen, the Netherlands, were included in a registry from January 2005 until December 2007.
From the beginning of 2005, in agreement with the referral sites (ambulance services and local hospitals), all pre-hospital patients with large infarctions (>15 mm cumulative ST-segment deviation) were referred for primary angioplasty. Pre-hospital patients with smaller infarctions (≤15 mm ST-segment deviation) were treated with either pre-hospital fibrinolysis or primary angioplasty depending on local agreements. Regular evaluation of the protocol demonstrated the feasibility and success of this primary angioplasty strategy, and pre-hospital fibrinolysis for smaller infarcts gradually decreased. Over time, primary angioplasty became the reperfusion strategy of choice for all patients. All patients who were presented to our intervention center were accepted for primary angioplasty. In case of an eligible patient, the resident of the intervention center was contacted either directly from the ambulance by paramedics or by the physician on call in the referral center. After acceptance, the catheterization laboratory personnel were directly activated to prepare the catheterization laboratory. If possible, the emergency room of the intervention center was bypassed and the patient was directly transported to the catheterization laboratory.
The Radboud University Medical Center is a university hospital and tertiary referral center. It is a high-volume center performing >1,000 angioplasty procedures/year, with 3 catheterization laboratories, surgical backup, and a team of experienced personnel with around-the-clock service. All operators perform >150 procedures/year. Before 2005, primary angioplasty was only performed in a selected group of patients (late myocardial infarction, shock, and in case of a contraindication for fibrinolysis) due to a successful pre-hospital fibrinolysis program with two-thirds of patients receiving fibrinolysis within 3 h of symptoms (10,11). Rescue angioplasty was performed in more than 100 patients/year.
The referral region is located in the east of the Netherlands and covers parts of the provinces of Gelderland and Brabant. It has a radius of about 48 miles with about 1,000,000 inhabitants. Five referral hospitals and 4 regional ambulance services are located in the referral region. Patients presenting with STEMI in a nonintervention center were treated with primary angioplasty, but were excluded from the current analysis of pre-hospital patients.
All ambulances in the region were equipped with the Lifepak 12 system (Physio Control, Redmond, Washington). Specially trained paramedics made a 12-lead electrocardiogram (ECG) in all patients with chest pain for <12 h of duration that did not resolve after sublingual nitroglycerin. This 12-lead pre-hospital ECG was interpreted by the paramedics with the help of a computer algorithm. In case of a clear-cut case of a patient with a large STEMI (>15 mm), paramedics contacted the cardiology resident of the intervention center to notify the catheterization laboratory before transportation. In other cases, pre-hospital fibrinolysis or primary percutaneous coronary intervention was chosen, often after initial contact with the cardiologist of the nearest hospital. All patients who were suspected of STEMI were given a loading dose of aspirin, clopidogrel, and unfractionated heparin before transport according to the prevailing guidelines at the time of inclusion.
Symptom time: time of symptom onset. 911 call time: time of 911 call. Diagnosis time = first medical contact time: time of first ECG on which STEMI was diagnosed. Door time: time of arrival at the intervention center. Balloon time: time of first balloon inflation or first device insertion. Total distance: the total distance was calculated with the use of an Internet route planner using postal codes. Direct transport: the distance between the pick-up address and the intervention center. Interhospital transport: the distance between the pick-up address and referral hospital plus the distance between the referral hospital and the intervention center.
Post-Procedural TIMI Flow Grade
The TIMI flow grade after the procedure was assessed by 2 experienced angiography analysts blinded to the study group of the patient. In case of discrepant findings, TIMI flow grade was determined by a third angiography analyst (12).
Information about the survival status of the patients was assessed on March 19 and 23, 2008, for the patients included in 2005 and 2006, and on May 5, 2009, for patients included in 2007 using the Municipal Personal Records Database.
Categorical variables are expressed as numbers with percentages and were compared by Pearson chi-square and Fisher's exact tests whenever appropriate. Normally distributed continuous variables are expressed as means with standard deviations and were compared using an independent t test. Continuous variables not normally distributed are expressed as medians with interquartile ranges (IQR) and were compared using the Mann-Whitney U test.
Multivariable binary logistic regression using forward stepwise inclusion was used to assess the relation between direct transport to the intervention center and the symptom-to-balloon time (divided at the median). Entered in the model were those variables (listed in Table 1) with a univariate association with the median symptom-to-balloon time (p < 0.10) and the baseline variables that differed (p < 0.10) between direct transport to the intervention center and interhospital transport. The symptom-to-911 call time was prospectively chosen to be included in the model.
To assess the relation between direct transport to the intervention center and the post-procedural TIMI flow grade 3 binary logistic regression was used with similar methodology as already described.
Long-term survival analysis was performed according to Kaplan and Meier, using the log-rank test for a univariate comparison between patients directly transported to the intervention center and patients referred through a nonintervention center.
Multivariable Cox regression analysis using forward stepwise inclusion was used to determine the association between direct transport to the intervention center and survival. Baseline characteristics (listed in Table 1) studied had a univariate association with survival (p < 0.10) or differed (p < 0.10) between direct transport and referral through a nonintervention center.
The symptom-to-balloon time and door-to-balloon time were prospectively chosen to be included in the multivariable models of post-procedural TIMI flow grade 3 and mortality.
In total, 581 pre-hospital patients with suspected STEMI were referred for primary angioplasty to the catheterization laboratory of the Radboud University Medical Center between January 2005 and the end of December 2007. All but 1 patient with ST-segment elevation on the pre-hospital ECG had STEMI. The other patient had an aortic dissection type A and died after the catheterization (direct transportation to the intervention center). No patients died during transport. Between first presentation and primary angioplasty, 14% of patients had an episode of ventricular fibrillation.
Direct transport to intervention center versus interhospital transport
Of the patients who first presented in the ambulance, 454 (78%) were directly transported to our intervention center and 127 (22%) were referred through a nonintervention center. The baseline and angiographic characteristics are shown in Table 1. Of the patients directly transported to the intervention center, 82% had a first-medical-contact-to-balloon time of <90 min and adhered to the time goal of the guidelines as compared to 23% of patients who were referred through a nonintervention center (p < 0.01). The corresponding median symptom-to-balloon times were 149 min (IQR 118 to 197 min) versus 219 min (IQR 178 to 315 min), p < 0.01 (Fig. 1). The symptom-to–911 call (Δ 7 min), 911 call–to-diagnosis (Δ 3 min) and the diagnosis-to-balloon times (Δ 49 min) were significantly shorter in patients who were directly transported to the intervention center. The door-to-balloon time was significantly longer in patients directly transported to the intervention center, 32 min (IQR 20 to 45 min) versus 24 min (IQR 17 to 36 min), p < 0.01. Patients directly transported to the intervention center more often had post-procedural TIMI flow grade 3 (92% vs. 84%, p = 0.03) and had lower mortality rates at 1 year (7% vs. 13%, p = 0.03) than patients after interhospital transport (Figs. 2 and 3).
Predictors of the symptom-to-balloon time
The symptom-to-balloon time was dichotomized at the median and was available in all patients. The univariable (p < 0.10) and multivariable characteristics associated with increased (above the median) symptom-to-balloon times are listed in Table 2.
Predictors of post-procedural TIMI flow grade 3
Post-procedural TIMI flow grade 3 was available in all patients. The univariable (p < 0.10) and multivariable characteristics associated with post-procedural TIMI flow grade 3 are listed in Table 3.
Follow-up at 1 year was complete for all but 1 patient (99.8%). The median total follow-up duration was 596 days (IQR 496 to 726 days). During follow-up, 64 patients (11%) had died. Univariable (p < 0.10) and multivariable predictors of long-term mortality are shown in Table 4.
To our knowledge, this is the largest prospective cohort of patients after pre-hospital diagnosis of STEMI comparing direct transport to an intervention center with interhospital transfer, and the first to suggest an effect of direct transport on long-term survival. Compared with interhospital transport, the rate of patients treated within the 90-min time frame of the guidelines more than tripled. Moreover, direct transport resulted in higher rates of post-procedural TIMI flow grade 3, a strong prognosticator of improved long-term survival. This was observed in a unique setting of a more than 10 years' experience with pre-hospital diagnosis of STEMI by trained paramedics, computerized ECG interpretation, and telephone consultation with a physician.
The observed time benefit of direct transportation is in concordance with previous reports also showing benefits with regard to improved time delays (4,5). Given the higher post-procedural TIMI flow grade 3 rate and the fact that guideline adherence has been associated with clinical outcome (13), this infrastructure should further improve outcome after STEMI. Several studies showed that pre-hospital diagnosis is associated with a greater use of reperfusion therapy (8), with a substantial reduction of treatment intervals, and with improved outcome in fibrinolysis (14) and primary angioplasty (4–6,8). In addition, in the current study, several interventions proven to be effective in reducing treatment intervals (3) are part of standard of care. Examples of these interventions are a single call from the ambulance to activate the catheterization laboratory, preparation of the catheterization laboratory while the patient is en route, and the fact that the intervention team had to be present within 20 min after being paged. Previous studies compared patients who presented at the referral center with patients directly transported to the intervention center after pre-hospital diagnosis. In contrast, all patients in our study underwent pre-hospital diagnosis of STEMI. The results of our study, therefore, underline the importance of direct referral to an intervention center, completely bypassing the emergency department if possible.
With regard to the safety, our data concur with the CAPTIM (Comparison of Angioplasty and Prehospital Thrombolysis in Acute Myocardial Infarction) trial (15) and DANAMI-2 (Danish Trial in Acute Myocardial Infarction 2) (16), both of which showed a low mortality rate during ambulance transport. Although not available in the current analysis, the reported sensitivity and specificity of a computer algorithm (often used in the pre-hospital setting) to detect STEMI are 50% to 60% and 96% to 100%, respectively (17). The relatively high mortality rate and the high prevalence of ventricular fibrillation in the current study is probably a reflection of the inclusion of substantial amount of older patients and patients presenting with large infarction. The apparent selection bias for older patients to be transferred through a referral hospital may be a consequence of the absence of an age limit for primary angioplasty. Doubt in the pre-hospital setting about the indication for primary angioplasty may have caused the preference for transport to the nearest referral center for additional evaluation. Moreover, patients may have been considered too sick for direct transport and may therefore have been presented at the nearest hospital.
Shorter symptom-to-balloon times were independently associated with higher rates of post-procedural TIMI flow grade 3 and patients directly transported to the intervention center had improved long-term survival. Direct transportation to the intervention center was independently associated with a shorter symptom-to-balloon time, a strong predictor of procedural success, which in turn is a prognosticator of long-term mortality.
In the current analysis, well-known independent predictors of outcome were confirmed. Pre-procedural TIMI flow grade and symptom-to-balloon time are well-established predictors of procedural success (18,19) and underline the importance of studies on antithrombotic pre-treatment and reducing treatment delays. Age, Killip class, and procedural outcome were confirmed to be independent prognosticators of long-term mortality (20,21). As procedural outcome is the only modifiable predictor, efforts should be made to perform primary angioplasty in high volume centers with experienced personnel (22). In previous transportation studies, the observed differences and sample sizes were too small to state conclusions with regard to clinical outcome. Our cohort study supports the previously observed trends of improved clinical outcome after direct transport to the intervention center. The fact that we observed higher post-procedural TIMI flow grade 3 rates after direct transport to the intervention center provides a plausible mechanistic explanation for the higher survival rate in this group.
Results from the current analysis are from an observational study in which the referral strategy for smaller infarctions was not uniform between referral sites and tended to change over time. Inherently, bias has been introduced with regard to the selection of either direct transport to the intervention center or interhospital transport.
This may have affected the observed clinical outcome differences, but the time intervals from 911 call to balloon will be merely unaffected. Although multivariable analysis was used, confounding of unknown and/or uncorrectable factors cannot be excluded. Although only established predictors of outcome were found in the Cox analysis, the model is overfit and should be interpreted conservatively.
Conclusions and Implications
Pre-hospital diagnosis of STEMI with direct notification of the catheterization laboratory and subsequent transportation to the intervention center is an attractive treatment strategy. It results in a more than tripling of the rate of patients treated within the 90-min time frame of the STEMI guideline compared with interhospital transport. Direct transportation to the intervention center resulted in a reduction of about 50 min in the diagnosis-to-balloon time. In case of uncertainty with regard to the diagnosis of STEMI or indication for reperfusion therapy, valuable time can be saved by direct transportation and evaluation in the intervention center, instead of opting for presentation at the nearest nonintervention center. With the symptom-to–911 call time representing one-third of the symptom-to-balloon time, renewed efforts should be made for patients' awareness (23). Time-to-balloon was an independent predictor of post-procedural TIMI flow grade 3, which underscores the need to reduce treatment delays in order to further improve clinical outcome after STEMI. Our data underscore that efforts should be made to organize a large-scale implementation of an infrastructure of pre-hospital diagnosis and direct transport to the intervention center, with early notification of the catheterization laboratory from the ambulance.
The authors thank Nick Clappers and Wessel Keuper for critical review of the manuscript. They gratefully acknowledge the efforts of Bart Kuipers and Kees van de Wulp with regard to data management.
- Abbreviations and Acronyms
- ST-segment elevation myocardial infarction
- Thrombolysis In Myocardial Infarction
- Received November 13, 2009.
- Revision received April 15, 2010.
- Accepted April 18, 2010.
- American College of Cardiology Foundation
- van de Werf F.,
- Ardissino D.,
- Betriu A.,
- et al.
- Gibson C.M.,
- Pride Y.B.,
- Frederick P.D.,
- et al.
- Terkelsen C.J.,
- Lassen J.F.,
- Norgaard B.L.,
- et al.
- Kudenchuk P.J.,
- Maynard C.,
- Cobb L.A.,
- et al.
- Rokos I.C.,
- French W.J.,
- Koenig W.J.,
- et al.
- Diercks D.B.,
- Kontos M.C.,
- Chen A.Y.,
- et al.
- Antman E.M.,
- Hand M.,
- Armstrong P.W.,
- et al.
- Daly C.,
- Clemens F.,
- Lopez-Sendon J.L.,
- et al.
- Kudenchuk P.J.,
- Ho H.T.,
- Weaver W.D.,
- et al.,
- MITI Project Investigators
- Brener S.J.,
- Moliterno D.J.,
- Aylward P.E.,
- et al.,
- APEX-AMI Investigators
- Halkin A.,
- Singh M.,
- Nikolsky E.,
- et al.
- De Luca G.,
- Suryapranata H.,
- van't Hof A.W.,
- et al.