Author + information
- Received July 27, 2007
- Revision received September 24, 2007
- Accepted October 9, 2007
- Published online February 1, 2008.
- David R. Holmes Jr, MD, FACC,
- Malcolm R. Bell, MBBS, FACC,
- Bernard J. Gersh, MB, ChB, DPhil, FACC,
- Charanjit S. Rihal, MD, MBA, FACC,
- Luis H. Haro, MD,
- Christine M. Bjerke, RN,
- Ryan J. Lennon, MS,
- Choon-Chern Lim, BS and
- Henry H. Ting, MD, MBA, FACC⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Henry H. Ting, Division of Cardiovascular Diseases, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905.
Objectives We implemented the Mayo Clinic ST-segment elevation myocardial infarction (STEMI) protocol and evaluated the timeliness of reperfusion therapy during off hours versus regular hours.
Background Patients with STEMI who present during off hours have longer door-to-balloon times and door-to-needle times.
Methods The Mayo STEMI protocol was implemented in May 2004 to optimize timeliness of reperfusion therapy for STEMI patients presenting to Saint Marys Hospital, a tertiary facility with on-site percutaneous coronary intervention (PCI), and for those presenting to 28 regional hospitals located up to 150 miles away from Saint Marys Hospital. We compared door-to-balloon times and door-to-needle times for 597 consecutive patients who presented during off hours (weekdays from 5 pm to 7 am and any time on weekends or holidays) versus regular hours (weekdays from 7 am to 5 pm). In 2003, prior to implementing the protocol, median door-to-balloon time at Saint Marys Hospital was 85 min during regular hours and 98 min during off hours.
Results Among 258 patients who presented to Saint Marys Hospital, median door-to-balloon time was 65 min during regular hours versus 74 min during off hours (p = 0.085). Among 105 patients transferred from regional hospitals for primary PCI, median door-to-balloon time was 118 min during regular hours versus 114 min during off hours (p = 0.15). Among 131 patients treated with fibrinolytic therapy at regional hospitals, median door-to-needle time was 21 min during regular hours versus 26 min during off hours (p = 0.067).
Conclusions The Mayo Clinic STEMI protocol demonstrates the rapid times that can be achieved through coordinated systems of care for STEMI patients presenting during off hours and regular hours.
Timely reperfusion therapy for patients with ST-segment elevation myocardial infarction (STEMI) is the cornerstone of treatment for this high risk group of patients (1–4). Guidelines and national quality initiatives have promulgated door-to-balloon time and door-to-needle time as metrics of quality (5–8). In the National Registry of Myocardial Infarction, patients with STEMI who presented during off hours (defined as 5 pm to 7 am on weekdays and anytime on weekends) as compared with regular hours (defined as 7 am to 5 pm on weekdays) had significantly longer door-to-balloon and door-to-needle times, as well as higher in-hospital mortality (9). Furthermore, the Myocardial Infarction Data Acquisition System registry recently showed that patients in New Jersey admitted with acute myocardial infarction on weekends as compared to weekdays were less likely to undergo cardiac catheterization and had higher mortality at 30 days (10). These studies demonstrated that patients who presented during off hours are less likely to undergo percutaneous coronary intervention (PCI) and experience longer hospital delays for administration of reperfusion therapy, and these delays in treatment may have consequences for adverse outcomes (11–14).
Bradley et al. (7) have developed strategies to improve timeliness of primary PCI, but it is not known whether these strategies are equally effective for patients with STEMI who present during regular hours and off hours. No such strategies exist for patients who received fibrinolytic therapy. We implemented and evaluated a novel system of care for patients with STEMI from a large, diverse geographic region including 28 hospitals up to 150 miles from a PCI center to determine if variations in door-to-balloon time and door-to-needle time during off hours on weekdays, weekends, and holidays could be minimized.
Data source and study sample
The Mayo Clinic STEMI protocol was developed to optimize timeliness of reperfusion therapy for a PCI center and a regional network of 28 hospitals, and details have been previously described (15). For patients with STEMI presenting to Saint Marys Hospital, the protocol was implemented in May 2004 and included the following strategies:
1. Prioritizing 12-lead electrocardiogram acquisition and physician interpretation within 10 min of hospital arrival for all patients with suspected acute coronary syndrome.
2. Emergency department activation of the cardiac catheterization laboratory without review or approval by cardiology.
3. Single call system to activate the entire cardiac catheterization team.
4. Catheterization laboratory to be fully operational within 30 min after activation.
5. Data collected prospectively using a computerized, Web-based database with feedback provided to staff within 24 to 48 h.
Strategies 2 to 5 represent 4 of the 6 strategies described by Bradley et al. (7). Prior to May 2004, these strategies were not routinely used at Saint Marys Hospital.
The regional STEMI protocol was implemented in December 2004 at 28 regional hospitals located in Minnesota, Wisconsin, and Iowa that were up to 150 miles away from Saint Marys Hospital. Regional hospitals were selected to participate if the total transfer time from the regional hospital to Saint Marys Hospital could be reliably achieved within 30 to 90 min (median 57 min), and if Saint Marys Hospital was the closest facility providing 24 h × 7 days PCI. The regional STEMI protocol included previous strategies adopted at Saint Marys Hospital and the following new strategies:
1. Standard order set and protocol used to select fibrinolytic therapy (for symptom onset <3 h) or PCI (for symptom onset >3 h) as primary reperfusion strategy and adjunctive medications.
2. Single phone call system for receiving cardiologist and regional physician to discuss case, activate air ambulance transfer, and activate the cardiac catheterization team.
3. A central communication center selected the fastest mode of transfer from 3 helicopters.
4. Helicopter “hot load” protocol with engine left running to minimize ground time to ≤10 min.
5. Saint Marys Hospital emergency department evaluation was bypassed.
Prior to December 2004, these strategies were not routinely used, and reperfusion choice and transfer were decided by individual physician preferences. As of December 2004, patients presenting to regional hospitals with symptom duration >3 h were transferred to Saint Marys Hospital for primary PCI. Patients presenting to regional hospitals within 3 h of symptom onset received full-dose fibrinolytic therapy unless they had a contraindication to fibrinolytic therapy or were considered to be at high clinical risk, such as those presenting with cardiogenic shock. All patients were transferred immediately after fibrinolytic drug administration to Saint Marys Hospital and evaluated by a cardiologist upon arrival. Patients underwent early rescue PCI for suspected failure to reperfuse or routine elective catheterization after 24 to 48 h if it was felt reperfusion had been successfully achieved with fibrinolytic therapy. Failed reperfusion after fibrinolytic therapy was defined as persistent chest discomfort or <70% resolution of ST-segment elevation 60 min after fibrinolytic drug administration. We did not use a facilitated PCI strategy defined as half- or full-dose fibrinolytic therapy followed by immediate catheterization.
Data collection and measures
This analysis compares door-to-balloon time, door-to-needle time, and outcomes for patients with STEMI who presented to the first hospital door during off hours (weekdays from 5 pm to 7 am and any time on weekends or official hospital holidays) versus regular hours (weekdays from 7 am to 5 pm). Door-to-balloon time was measured as the time from first hospital arrival either at Saint Marys Hospital or at a regional hospital to the time of any therapeutic device (balloon, stent, or thrombectomy catheter) in the infarct-related artery. Door-to-needle time was defined by time from first hospital arrival to administration of fibrinolytic therapy. For Saint Marys Hospital patients, pre-specified time intervals that comprised total door-to-balloon time were prospectively collected including door-to-electrocardiogram, electrocardiogram-to-catheterization laboratory activation, catheterization laboratory activation-to-catheterization laboratory arrival, and catheterization laboratory arrival-to-balloon. For regional hospital patients, pre-specified time intervals that comprise total door-to-balloon time were also collected including door1-to-electrocardiogram, electrocardiogram-to-protocol activation, protocol activation-to-door2, and door2-to-balloon. Door1 was defined as the regional hospital and door2 was defined as Saint Marys Hospital. For regional hospital patients, pre-specified time intervals that comprise total door-to-needle time were collected including door1-to-electrocardiogram and electrocardiogram-to-needle.
Door-to-balloon times were also prospectively collected for 60 patients with STEMI who presented to Saint Marys Hospital in 2003 before the Mayo Clinic STEMI protocol was implemented. We compared the door-to-balloon times for patients with STEMI who presented to Saint Marys Hospital before and after implementation of the protocol in May 2004.
Cardiogenic shock was defined as persistence of systolic blood pressure less than 85 mm Hg unresponsive to fluid challenge and the requirement for vasopressors or placement of an intra-aortic balloon pump. Recurrent myocardial infarction within 24 h of qualifying STEMI required typical ischemic discomfort ≥20 min and new or recurrent ST-segment elevation ≥0.10 mV in ≥2 continuous leads or new left bundle branch block. Recurrent myocardial infarction after 24 h of the qualifying STEMI required typical ischemic discomfort ≥20 min in addition to one of the following:
1. New or recurrent ST-segment elevation or new left bundle branch block, or
2. Both an increase in creatine kinase-myocardial band (CK-MB) isoenzymes of at least 50% over previous value and documentation that this level has been decreasing prior to the onset of the suspected recurrent myocardial infarction (MI). Biomarkers were drawn at the time of first hospital arrival and every 8 h for 24 h; following this, they were drawn at 48 and 72 h.
Continuous variables are summarized as mean ± standard deviation for symmetric distributions and median (25th, 75th percentiles) for skewed variables. Discrete variables are presented as frequency (percentage). Group differences were tested using 1-way analysis of variance, the Kruskal-Wallis rank-sum test, and Pearson’s chi-square test, respectively. Ninety-five percent confidence intervals for in-hospital event rates were computed using exact binomial methods. Kaplan-Meier estimation was used for time-to-event type variables such as time to treatment and follow-up mortality with group differences tested by the log-rank test. Confidence intervals for survival estimates were calculated using the log transformation.
From May 2004 to December 2006, 597 patients presented with new or presumed new ST-segment elevation or left bundle branch block. Exclusions included 40 patients who refused research consent and 63 additional patients when final clinical diagnoses other than STEMI were identified such as myocarditis, pericarditis, apical ballooning syndrome, and left ventricular aneurysm or hypertrophy (Fig. 1). The study sample included 258 patients who presented to Saint Marys Hospital and underwent primary PCI; 105 patients who presented >3 h from symptom onset to a regional hospital and were transferred to Saint Marys Hospital for primary PCI; and 131 patients who presented within 3 h from symptom onset to a regional hospital, were treated with full-dose fibrinolytic therapy, and then transferred to Saint Marys Hospital. Of these patients, 198 (40%) were seen and treated during regular hours and 296 (60%) during off hours. Patients seen during off hours were slightly more likely to be diabetic (20% vs. 12%, p = 0.029). There were no other significant differences in baseline characteristics between patients presenting during regular hours versus off hours (Table 1). The mean age was approximately 64 years, inferior infarction predominated, being present in approximately 55% and only 7% to 8% had cardiogenic shock.
Reperfusion therapy time-to-treatment
The time from symptom onset to first hospital arrival did not vary significantly for regular hours versus off hours presentation (Table 1) and was relatively short at 88 min versus 92 min, respectively. For the patients presenting to Saint Marys Hospital, median door-to-balloon time was 65 min during regular hours and 74 min during off hours (p = 0.085) (Fig. 2A). Among patients with STEMI transferred for primary PCI from a regional hospital, overall door-to-balloon times were longer because of the time required for transfer, but median door-to-balloon time was not statistically different during regular hours (118 min) versus off hours (114 min), (p = 0.15) (Fig. 2B). Among patients with STEMI treated with full-dose fibrinolytic therapy at a regional hospital prior to transfer, median door-to-needle time was slightly shorter during regular hours versus off hours, 21.5 min versus 26.0 min, but this did not quite reach statistical significance (p = 0.067) (Fig. 2C).
Pre-specified time intervals that comprised total door-to-balloon time were compared for patients who presented during regular hours versus off hours for Saint Marys Hospital (Fig. 3A) and regional hospitals (Fig. 3B). For the Saint Marys Hospital group, the time interval from catheterization laboratory activation to catheterization laboratory arrival was increased during off hours versus regular hours (24.0 min vs. 14.0 min, p < 0.001) and all the other time intervals were similar. For the regional hospital PCI group, the time interval from door1 to electrocardiogram was slightly increased during off hours versus regular hours (7.5 min vs. 4.0 min, p = 0.039) and all the other time intervals were similar. Pre-specified time intervals that comprise total door-to-needle time were compared for patients who presented to a regional hospital during regular hours versus off hours (Fig. 3C). The electrocardiogram to needle time was slightly increased during off hours as compared to regular hours but this did not reach statistical significance (19.0 min vs. 15.5 min, p = 0.083).
Time to treatment of Saint Marys Hospital patients versus control patients
Door-to-balloon times were significantly decreased for patients who presented to Saint Marys Hospital after implementation of the Mayo Clinic STEMI protocol in May 2004 as compared to those patients who presented to Saint Marys Hospital in 2003 before implementation of the protocol. During regular hours, median door-to-balloon time was 65 (interquartile range [IQR] 50 to 90) min after implementation of the protocol versus 85 (IQR 73 to 111) min before implementation, (p = 0.019) (Fig. 4A). During off hours, median door-to-balloon time was 74 (IQR 60 to 93) min after implementation of the protocol versus 98 (IQR 70 to 125) min before implementation, (p = 0.005) (Fig. 4B).
In-hospital and follow-up outcomes are summarized in Table 2. There were no significant differences in frequency of mortality or recurrent MI during hospitalization. In-hospital death occurred in 4.5% versus 6.1% for regular hours versus off hours (p = 0.46), and the combined end point of death or MI occurred in 7.6% and 8.8%, respectively. The median follow-up time for the entire group was 184 days, and Kaplan-Meier estimates demonstrated no difference in mortality at follow-up for patients presenting during regular hours (9.8%) versus off hours (8.6%) (p = 0.69) (Fig. 5).
Our study found that door-to-balloon times can be significantly improved for patients with STEMI who presented to Saint Marys Hospital, a PCI center, during off hours and regular hours by implementing simple strategies as compared with door-to-balloon times before implementation of such strategies. However, even though faster door-to-balloon times were achieved after implementation of the Mayo Clinic STEMI protocol, patients who presented to Saint Marys Hospital still had slightly longer median door-to-balloon times during off hours (74 min) as compared with during regular hours (65 min) (p = 0.085). This difference is attributable to a longer time interval from catheterization laboratory activation to catheterization laboratory arrival during off hours. Among patients with STEMI who presented to 28 regional hospitals up to 150 miles away, we implemented novel strategies to coordinate systems of care and patient transfer, and we were able to achieve rapid door-to-balloon times and door-to-needle times regardless of presentation during off hours versus regular hours. Because the median total transfer time from door1 to door2 was 57 min, this buffer provided sufficient time for the on-call catheterization team to be activated and respond during off hours.
Strategies to improve door-to-balloon times
Our Mayo Clinic STEMI protocol implemented 4 strategies from the Bradley et al. paper (7), namely: 1) having the emergency medicine physician activate the catheterization laboratory; 2) using a single call system to activate the catheterization laboratory; 3) expecting the catheterization laboratory staff to arrival within 20 to 30 min of activation; and 4) using real-time data for performance feedback. However, by allowing up to 20 to 30 min for the on-call catheterization team to arrive during off hours, a finite delay is embedded into the processes of care during off hours. This time interval for the on-call catheterization team to respond after activation accounted for the bulk of the difference in door-to-balloon time for patients with STEMI who presented to Saint Marys Hospital during off hours (74 min) versus regular hours (65 min). To overcome this delay in reperfusion therapy, we may need to consider adopting one or both of the remaining strategies from Bradley et al. (7) that are most complex and costly to implement: 1) using pre-hospital electrocardiograms to activate the catheterization laboratory while patient is en route, and 2) having an interventional team always on site 24 h × 7 days.
Rapid initiation of reperfusion therapy for STEMI has been found to limit infarct size and improve survival (1,2). With primary PCI, several authors have addressed the issue of time dependency and mortality (16–24). McNamara et al. (22) found that a door-to-balloon time of <90 min was associated with an in-hospital mortality of 3.0% compared with mortalities of 4.2%, 5.7%, and 7.4% for door-to-balloon times of 91 to 120 min, 121 to 150 min, and >150 min, respectively. In patients in whom fibrinolytic therapy is considered, there is a striking time dependency on survival that is sometimes referred to as the “golden hour” of reperfusion (25,26). Clearly fibrinolytic therapy has the greatest benefit for myocardial salvage and mortality reduction for patients who present early but the magnitude of these benefits decreases over time.
Timeliness of reperfusion therapy during off hours
These studies among others have led to the evidence-based guidelines that emphasize timeliness of reperfusion therapy as a metric of quality (1,2,5–8). However, effective implementation and uniform adoption in “real world” practices have been difficult, and many institutions fail to meet these benchmarks with potentially adverse consequences on patient outcome. Magid et al. (9) found: 1) door-to-balloon times were longer in patients presenting during off hours; 2) this pattern was consistent across all hospital types; and 3) patients presenting off hours had a higher in-hospital mortality as a consequence of these delays in reperfusion therapy. The reasons for these delays are often “system” based and related to the scope and intensity of how care is delivered and staffing models during off hours. Kostis et al. (10) evaluated these issues in 231,164 patients admitted with acute MI throughout New Jersey from 1987 to 2002. They found that despite the lack of significant differences in demographics and comorbid conditions, patients admitted on weekends had higher mortality and lower use of invasive cardiac procedures as compared to weekdays. As hospitals and national quality improvement efforts seek to improve performance for rapid and appropriate reperfusion therapy in patients with STEMI, we should strive to develop systems of care in real world practices that are designed to minimize delays in door-to-balloon time and door-to-needle time regardless of whether the patient presents during regular hours or off hours or which hospital types the patients initially presented to.
This is a study of a single regional system of care, although it included a variety of hospital types across a large geographic region. This was not a trial to compare outcomes of different reperfusion strategies, and the number of events is relatively small. The results are unadjusted, and multivariable analysis was not performed.
Our study implemented and evaluated a protocol to improve timeliness of reperfusion therapy for a PCI center and regional network of 28 hospitals. We found that patients with STEMI who presented during off hours to a PCI center continued to be at risk for slight hospital delays in door-to-balloon time, even though improvements in times were seen during off hours and regular hours. This gap represents an area for continued research to improve the quality of reperfusion therapy. Our protocol achieved rapid and similar door-to-balloon and door-to-needle times for patients who initially presented to a regional hospital and were subsequently transferred to a PCI center.
This study was supported by the Division of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota. Douglas Weaver, MD, acted as Guest Editor for this paper.
- Abbreviations and Acronyms
- coronary artery bypass graft
- congestive heart failure
- interquartile range
- myocardial infarction
- percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- Received July 27, 2007.
- Revision received September 24, 2007.
- Accepted October 9, 2007.
- American College of Cardiology Foundation
- Antman E.M.,
- Anbe D.T.,
- Armstrong P.W.,
- et al.
- Van de Werf F.,
- Ardissino D.,
- Betriu A.,
- et al.,
- task force on the management of acute myocardial infarction of the European Society of Cardiology
- Van de Werf F.,
- Baim D.S.
- McNamara R.L.,
- Herrin J.,
- Bradley E.H.,
- et al.
- Nallamothu B.K.,
- Bates E.R.,
- Herrin J.,
- et al.
- Waters R.E.,
- Singh K.P.,
- Roe M.T.,
- et al.
- Ting H.H.,
- Rihal C.S.,
- Gersh B.J.,
- et al.
- Berger P.G.,
- Ellis S.G.,
- Holmes D.R. Jr.,
- et al.
- De Luca G.,
- Suryapranata H.,
- Ottervanger J.P.,
- Antman E.M.
- Boersma E.,
- Primary Coronary Angioplasty vs. Thrombolysis Trialists
- McNamara R.L.,
- Wang Y.,
- Herrin J.,
- et al.
- Pinto D.S.,
- Kirtane A.J.,
- Nallamother B.K.,
- et al.
- Steg P.G.,
- Bonnefoy E.,
- Chabaud S.,
- et al.