Methods

The National Cardiovascular Data Registry Acute Coronary Treatment Intervention Outcomes Network Registry–Get With The Guidelines (ACTION Registry–GWTG) is an initiative of the American College of Cardiology and the American Heart Association, evaluating consecutive patients with acute myocardial infarction (MI) treated at participating hospitals across the United States. The institutional review board at each of the hospitals approved participation in the National Cardiovascular Data Registry ACTION Registry–GWTG. ACTION Registry–GWTG serves as a hospital data collection and evaluation mechanism for American Heart Association’s Mission: Lifeline. The Duke Clinical Research Institute serves as the analytical center and has an agreement to analyze the aggregate de-identified data for research purposes.

There were 113,305 patients with STEMI from 597 centers enrolled from January 2007 to March 2011. STEMI was defined as persistent ST-segment elevation ≥1 mm in ≥2 contiguous leads, a new or presumed new left bundle branch block, or an isolated posterior MI. Patients were excluded from the analysis sequentially: centers using the limited data collection form because important clinical factors were not collected (n = 8,152), patients enrolled in hospitals without PCI capability (n = 3,726 with 71 centers), transfer-in patients (n = 35,352), patient’s mode of transport to the first facility was EMS or unknown (n = 40,764), patients with STEMI equivalent first noted on subsequent electrocardiogram (ECG) or unknown (n = 2,676), patients not undergoing primary PCI or with missing primary PCI status (n = 4,630), patients receiving fibrinolytics (n = 34), patients with nonsystem reason for delay in PCI (difficult vascular access, cardiac arrest and/or need for intubation before PCI, patient delay in providing consent, or difficulty in crossing the culprit lesion during PCI, n = 1,711), D2B time missing or >12 hours (n = 133), or patient’s symptom onset time to hospital presentation was zero, negative, missing, or >12 hours (n = 2,748). Thus, the final analysis population consisted of 13,379 patients from 432 participating hospitals.

Definitions for data elements in the ACTION Registry–GWTG database are available online ( http://www.ncdr.com/WebNCDR/NCDRDocuments/ACTIONv2CodersDataDictionary_2.1.1.pdf ). Self-transport was defined as any transportation that did not involve EMS; this included transport with taxis or public transportation, transport in a private car operated by self or others, and walking to the hospital. Off-hour presentation was defined as occurring from 7 p.m. to 7 a.m. during weekdays and on weekends. D2B time was defined as hospital arrival to primary PCI and was further dichotomized into D2B time ≤ or >90 minutes. The year of hospital presentation was defined by the year of arrival to the hospital (e.g., 2007 to 2011). In-hospital major bleeding was defined as an absolute hemoglobin (Hgb) decrease of ≥4 g/dl (baseline to nadir), intracranial hemorrhage, documented or suspected retroperitoneal bleed, any red cell blood transfusion with baseline Hgb ≥9 g/dl, or any red cell transfusion with Hgb <9 g/dl, and a suspected bleeding event. Given that most patients who undergo coronary artery bypass surgery (CABG) receive blood transfusions related to the surgery, bleeding events were considered only if they occurred before CABG.

Baseline demographics, medical history, presentation variables, hospital characteristics, time intervals, and in-hospital clinical outcomes were compared between D2B > and ≤90 minutes. Categorical variables are expressed as number of patients and percentiles; continuous variables are expressed as median (twenty-fifth, seventy-fifth percentiles) or mean ± SD. Chi-square test was used to compare categorical variables, and Wilcoxon rank-sum 2-sample test was used to compare continuous variables. Temporal trends of D2B ≤90 minutes was displayed in a bar graph, in which logistic regression analysis was used to test for linear trend by fitting the year of hospital presentation as an ordinal covariate.

To explore factors associated with prolonged D2B (>90 minutes), logistic generalized estimating equation method with an exchangeable working correlation matrix was used to account for correlations among clustered responses because patients within a hospital are more likely to have similar responses relative to patients at other hospitals (i.e., within-center correlation for responses). This method produces estimates similar to those from ordinary logistic regression, but variances are adjusted for the correlation of outcomes within a hospital. Continuous variables were tested for linearity and plotted against rates of prolonged D2B to create dichotomous cut-off points when applicable. Cut-off points were considered when the relation between the variable and prolonged D2B became flat or nonlinear and finalized once determined to be clinically appropriate. Only statistically significant factors were included in the model. Furthermore, the relation between in-hospital all-cause mortality and major bleeding and prolonged D2B was examined using logistic generalized estimating equation method adjusted for covariates from the previously validated ACTION Registry–GWTG in-hospital mortality and major bleeding models. Covariates included age, gender, race, weight, heart failure at presentation, shock at presentation, heart rate, systolic blood pressure, electrocardiographic findings (ST-segment elevation, ST-segment depression, or transient ST-segment elevation vs T-wave inversion or no electrocardiographic changes), hypertension, diabetes mellitus, previous peripheral artery disease, current/recent smoker, dyslipidemia, previous MI, previous PCI, previous CABG, previous heart failure, previous stroke, baseline Hgb, baseline serum creatinine, baseline troponin ratio (times the upper limit of normal), and home medications including aspirin, clopidogrel, warfarin, β blocker, angiotensin-converting enzyme inhibitor, angiotensin receptor blocker, aldosterone-blocking agent, statin, nonstatin lipid-lowering agent. Adjusted associations for outcomes are displayed as odds ratios (95% confidence intervals). All analyses were performed using SAS software (version 9.2; SAS Institute, Cary, North Carolina).

Results

The median (twenty-fifth, seventy-fifth percentiles) D2B time for the entire study cohort (n = 13,379) was 72 minutes (58, 86), and 81% of patients had a D2B time of ≤90 minutes. During the study period, there was a significant increase in the percentage of patients achieving a D2B time ≤90 minutes ( Figure 1 ).

Percentage of patients achieving D2B time ≤90 minutes from 2007 through 2011.

Patients with D2B time >90 minutes were more likely to be women, African-American, and to have a history of hypertension, diabetes mellitus, renal failure requiring dialysis, CABG, and cerebrovascular disease than patients with D2B time ≤90 minutes ( Table 1 ). Patients with D2B time >90 minutes were more likely to present with evidence of heart failure, had higher initial troponin levels, and more commonly presented during off-hours. Median time to initial ECG was significantly longer in patients with D2B time >90 minutes (11 [5, 21] vs 5 [2, 9] minutes, p <0.001). Obtaining the initial ECG within 10 minutes of arrival occurred in 47.1% of those with D2B time >90 minutes compared with 79.5% of patients with D2B time ≤90 minutes (p <0.0001).

Stay updated, free articles. Join our Telegram channel

Join