False Activation of the Cardiac Catheterization Laboratory for Primary PCI
Published Online: August 13, 2013
Geoffrey D. Barnes, MD; Alexander Katz, MD; Jeffrey S. Desmond, MD; Steven L. Kronick, MD, MS; Jamie Beach, RN; Stanley J. Chetcuti, MD; Eric R. Bates, MD; and Hitinder S. Gurm, MD
The American College of Cardiology/American Heart Association (ACC/AHA) guidelines for management of ST-elevation myocardial infarction (STEMI) recommend that patients should be treated with primary percutaneous coronary intervention (P-PCI) within 90 minutes of first medical contact.1,2 A national effort was started by the ACC in 2006 to reduce door-to-balloon (D2B) times, defined as the time from patient arrival at the hospital emergency department (ED) to reperfusion in the cardiac catheterization laboratory (CCL).3 This process included encouraging emergency medical service (EMS) providers (primarily paramedics) and ED physicians to activate the CCL prior to consultation with a staff cardiologist.3-6 Hospital D2B times are publically reported, so clinicians may feel pressure to rush and activate the CCL before a review of the patient’s presenting condition can be completed, and occasionally CCL may be activated when the patient does not have STEMI. While there is increasing awareness of false CCL activation, the change in false CCL activation over time has not been assessed since the ACC initiated the D2B Alliance campaign.
Goals of This Investigation
Our objective was to describe the trends in the rate of false CCL activation and D2B time at our institution over the 5 years following the national D2B initiative. We hypothesized that as D2B times have fallen, the rate of false CCL activations has increased.
Selection of Patients and Setting
Between January 1, 2007, and December 31, 2011, all adult patients with suspected STEMI that presented to the University of Michigan Health System ED were reviewed. This study was approved by the institutional review board at the University of Michigan, a large academic suburban hospital. Data were initially collected retrospectively in a quality improvement database by 1 author (JB) and then verified through the group paging system by another author (GDB). Additional data were abstracted from the chart using a predefined set of data elements by 2 authors (GDB, AK). Selected cases werereviewed by both abstractors, with any discrepancy resolved by consensus.
Study Design and Outcome Measures
For the purpose of this study, a false activation is defined as one in which the CCL was activated by EMS or the ED for new STEMI or left bundle branch block (LBBB), but the patient was determined by the interventional cardiologist to not require emergent transfer from the ED to the CCL for P-PCI. To be as inclusive as possible for other reasons for emergent activation of the CCL, patients deemed too unstable to transfer to the CCL, who refused cardiac catheterization, who presented with cardiac arrest, who died prior to arrival in the CCL, or who were taken to the CCL emergently for any other indication, were defined as true CCL activations. Findings in the CCL did not influence the determination of a true or false CCL activation. The decision to not take a patient to the CCL was made initially by the on-call interventional cardiology fellow and verified in all cases by the interventional cardiology attending.
The primary outcome was the rate of false CCL activation, defined as the number of false CCL activations divided by the total number of CCL activations. Discharge diagnoses, peak serum troponin levels, and results of any cardiac catheterization during the index hospitalization were reviewed for all patients and compared with the initial decision to proceed or not proceed emergently to the CCL. Patients were classified by the year they presented to the ED and by CCL activation location (pre-hospital or ED). Other variables examined included age and gender. The D2B time was calculated in minutes as the time from ED arrival until initial device deployment for reperfusion, when appropriate.
Logistic regression analysis was conducted to demonstrate the relationship between year of presentation, location of CCL activation, and true/false CCL activation, as well as D2B time and year of presentation. Adjusted odds ratios were calculated by incorporating the patient’s age and gender into the logistic regression analysis, but these did not impact the final model and so are not presented. Model fit was assessed with the Hosmer-Lemeshow goodness of fit test.7 Logistic regression analysis was also conducted to demonstrate the relationship between year of presentation and location of CCL activation independent of true/false CCL activation status. Statistical analyses were conducted with the use of SPSS version 19 (Armonk, New York).
Between January 1, 2007, and December 31, 2011, the CCL was activated 717 times for possible STEMI. The average patient age was 61.4 (standard deviation [SD] 14.4) years and 72.4% were male. Pre-hospital CCL activations ranged from 33% in 2007 to 52% in 2011. For those patients who received P-PCI, the median D2B time decreased from 67 (SD 37.2) minutes in 2007 to 55 (SD 33.6) minutes in 2011 (3.74 min/year, 95% confidence interval [CI] 1.36-6.11 min/year, P = .002, Figure).
True and False CCL Activations
There were 513 (72%) true CCL activations and 204 (28%) false CCL activations. Of the true CCL activations, 73 (14%) patients were never taken to the CCL due to pulseless electrical activity, ventricular fibrillation, hemodynamic instability, refusal, or death. Yearly false CCL activation rates ranged from 15% in 2007 to 40% in 2011, a 37% per year increase in the odds of false CCL activation (95% CI 21-56%, P <.001, Figure). Of the 717 CCL activations, 315 (44%) were pre-hospital activations (33% in 2007 to 52% in 2011, Table). False CCL activations were similar for ED versus pre-hospital location (odds ratio 1.12, 95% CI0.62-2.07, P = .71).
Reasons for False CCL Activation
PDF is available on the last page.