The percentage of patients with primary percutaneous coronary intervention (PCI) with door-to-balloon (D2B) times ≤90 minutes is used as a hospital performance measure for public reporting. Patients can be excluded from reporting for nonsystem-related delays. How exclusions impact D2B time reporting at the hospital level is unknown. The percentage of patients having nonsystem delays for primary PCI at the hospital level was calculated using data from the Acute Coronary Treatment Intervention Outcomes Network Registry-Get with the Guidelines Registry. Hospitals were categorized based on tertiles of percentage of excluded patients: low, ≤7.1%; intermediate, >7.1% to 11.2%; and high, >11.2%. From January 1, 2007, to March 31, 2011, 43,909 patients from 294 hospitals were included. The percentage of exclusions differed substantially among hospitals (0% to 68%, median 9.2% [interquartile range 5.6% to 13.5%]). Exclusion reasons included vascular access difficulty (7.6%), cardiac arrest/intubation (38%), and PCI procedural difficulties (20%). Including patients with nonsystem delays significantly increased D2B times by ≤2 minutes for each group. The effect was larger on the proportion of patients having a D2B ≤90 minutes (low 83.6% to 85%, intermediate 82.9% to 86.3%, high 82% to 87.5%, p <0.001, for all). If a criterion of having ≥90% of patients with D2B ≤90 minutes was used, excluding patients with nonsystem delays significantly increased the proportion of patients meeting this goal for each group: low, 28% to 37%; intermediate, 17.7% to 37.5%; and high, 14% to 52% (all p <0.01). In conclusion, the proportion of patients excluded from D2B reporting varies substantially among hospitals. This has a greater impact on percentage of patients with D2B time ≤90 minutes than on median D2B times.
In patients with ST-segment elevation myocardial infarction (STEMI), faster door-to-balloon (D2B) times have been associated with lower mortality. Substantial efforts from the American College of Cardiology and their D2B Alliance program, and subsequently, the American Heart Association Mission: Lifeline program, have substantially reduced times for primary percutaneous coronary intervention (PCI) from times that often exceeding 120 minutes to times now well below the current guideline of ≤90 minutes. In conjunction with improvement in D2B times, the Centers for Medicaid & Medicare Services (CMS), in 2006, and the National Cardiovascular Data Registry’s CathPCI Registry in 2009, expanded exclusions to include patients with prolonged D2B times as a result of nonsystem delays. The degree to which these exclusions impact reported D2B times has not been well assessed. Therefore, using the National Cardiovascular Data Registry’s Acute Coronary Treatment Intervention Outcomes Network Registry-Get with the Guidelines Registry (ACTION Registry-GWTG), we analyzed the frequency of patient exclusions as a result of nonsystem delays and its impact on D2B times.
Methods
The ACTION Registry-GWTG is a database containing records from >600 participating hospitals. The registry uses a standardized data set with written definitions, has requirements in place to ensure uniform data entry and transmission, and is subject to data quality checks. Participating hospitals collect data through retrospective chart review using standardized data collection tools that do not require direct contact with individual patients. This registry was either approved by an institutional review board or considered quality assurance data and not subject to institutional review board approval based on individual site determinations. Data definitions for the elements, including the definitions for the clinical endpoints of in-hospital mortality, non-STEMI, STEMI, heart failure, and cardiogenic shock, are available online ( http://www.ncdr.com/WebNCDR/NCDRDocuments/ACTION_v2_CodersDataDictionary_2.1.1.pdf ). National Cardiovascular Data Registry has validated the data through a Data Quality Program to ensure accuracy and data completeness.
This study included 113,305 patients with STEMI admitted to 597 PCI-capable hospitals from January 2007 to March 2011. Patients were excluded if the limited data form was used (n = 8,152), STEMI or STEMI equivalent was first noted on a subsequent electrocardiogram (n = 10,961), left bundle branch block or isolated posterior MI was present on the initial electrocardiogram (n = 3,309), primary PCI was not performed (n = 25,175), thrombolytic therapy was administered (n = 734), they were transferred-in or transferred-out (n = 18,010), or D2B time was >12 hours or missing (n = 326). Of the remaining 46,638 patients, 2,729 were excluded from 165 hospitals that treated <36 patients during the study period. The remaining 43,909 patients from 294 hospitals comprised the study population.
The study population was separated into 3 groups (low, intermediate, and high exclusion rates) based on tertiles of the proportion of patients who had documented reasons for nonsystem delays, which included the following: (1) difficult vascular access, (2) cardiac arrest and/or need for intubation before PCI, (3) patient delays in providing consent for procedure, (4) difficulty crossing the culprit lesion during the PCI procedure, and (5) other.
The ACTION Registry-GWTG serves as the hospital data collection and evaluation mechanism for the American Heart Association Mission: Lifeline STEMI program. The Duke Clinical Research Institute serves as the data analysis center and has an agreement to analyze the aggregate de-identified data for research purposes.
Continuous variables are presented as medians with 25th and 75th percentiles, and categorical variables are expressed as frequencies with percentages. Kruskal-Wallis test was used to compare continuous variables and Mantel-Haenszel chi-square test was used to compare categorical variables. All analyses were conducted using SAS 9.2 (SAS Institute Inc., Cary, North Carolina).
Results
A total of 43,909 patients from 294 hospitals comprised the study population. The proportion of patients with STEMI who had nonsystem delay exclusions among hospitals was a median of 9.2% and ranged from 0% to 68% (interquartile range 5.6% to 13.5%; Figure 1 ).
Among patients with nonsystem delay reasons, the most common reason for exclusion was cardiac arrest and/or need for intubation before PCI (38.4%), while difficulty crossing the culprit lesion (19.8%), difficult vascular access (7.6%), and delays in providing consent for the procedure (5.4%) were the next most frequent documented reasons for nonsystem delays.
Hospitals in the low exclusion group (n = 100) excluded 4.0% of patients (range 0% to 7.1%), compared with 9.1% (range >7.1 to 11.3%) in the intermediate (n = 96) and 16.8% (range >11.3% to 68%) in the high exclusion groups (n = 98). Absolute differences in age and risk factors were small and likely not clinically different ( Table 1 ). The frequency of cardiogenic shock was less common in the hospitals in the lower exclusion groups (6.9% vs 7.0% vs 7.6%). There were no significant differences in hospital annual MI volume, annual STEMI volume, hospital bed size, hospital location (rural or nonrural), and teaching hospital status (academic or nonacademic) among the 3 exclusion groups ( Table 2 ).
Variable | Non-Systems Delays | ||||
|---|---|---|---|---|---|
| All Patients (n = 43,909) | Low (N = 13,441) | Intermediate (N = 18,090) | High (N = 12,738) | p Value | |
| Age, years | 60 (51, 69) | 59 (51, 69) | 59 (51, 69) | 60 (52, 70) | <0.0001 |
| Male | 72.1% | 71.9% | 72.5% | 71.7% | 0.26 |
| Hypertension | 61.0% | 61.4% | 59.8% | 62.2% | 0.0001 |
| Diabetes mellitus | 21.2% | 21.0% | 20.6% | 22.2% | 0.003 |
| Current/recent smoker | 45.3% | 45.9% | 45.5% | 44.3% | 0.024 |
| Dyslipidemia | 51.9% | 51.9% | 51.7% | 52.2% | 0.70 |
| Peripheral arterial disease | 4.9% | 5.1% | 4.6% | 5.1% | 0.09 |
| Prior myocardial infarction | 18.9% | 18.9% | 18.6% | 19.4% | 0.19 |
| Prior percutaneous coronary intervention | 21.2% | 21.5% | 21.4% | 20.5% | 0.13 |
| Prior bypass surgery | 5.7% | 5.9% | 5.6% | 5.8% | 0.45 |
| Prior heart failure | 3.6% | 3.7% | 3.1% | 4.0% | 0.0002 |
| Prior stroke | 4.1% | 4.2% | 3.8% | 4.6% | 0.003 |
| At presentation | |||||
| Heart rate (beats/min) | 77 (64, 91) | 77 (64, 91) | 77 (64, 90) | 77 (64, 91) | 0.28 |
| Systolic blood pressure (mm Hg) | 138 (118, 159) | 139 (118, 159) | 138 (118, 159) | 138 (117, 160) | 0.75 |
| Cardiogenic shock | 6.8% | 5.9% | 7.0% | 7.6% | <0.0001 |
| Signs of heart failure | 7.2% | 8.0% | 6.9% | 6.6% | <0.0001 |
| Hemoglobin (g/dL) | 14.5 (13.3, 15.6) | 14.4 (13.2, 15.5) | 14.5 (13.3, 15.6) | 14.5 (13.3, 15.6) | <0.0001 |
| Serum creatinine (mg/dL) | 1.0 (0.9, 1.2) | 1.0 (0.9, 1.2) | 1.0 (0.9, 1.2) | 1.0 (0.9, 1.2) | 0.47 |
| Creatinine clearance (mL/min) | 88 (65, 114) | 88 (65, 115) | 89 (65, 115) | 87 (64, 113) | <0.0001 |
| Variable | All Hospitals (n = 294) | Low Proportion of Non-Systems Delays (N = 100) | Intermediate Proportion of Non-Systems Delays (N = 96) | High Proportion of Non-Systems Delays (N = 98) | p Value |
|---|---|---|---|---|---|
| Hospital type | 0.49 | ||||
| Nonacademic | 77.9% | 76.0% | 82.3% | 75.5% | |
| Academic | 21.8% | 24.0% | 17.7% | 22.5% | |
| Hospital region | 0.15 | ||||
| West | 16.3% | 10.0% | 14.6% | 24.4% | |
| Northeast | 8.9% | 11.0% | 9.4% | 6.1% | |
| Midwest | 33.3% | 32.0% | 36.4% | 31.6% | |
| South | 41.5% | 47.0% | 39.6% | 37.8% | |
| Number of hospital beds | 343 (244, 493) | 347 (258, 497) | 346 (237, 513) | 330 (242, 457) | 0.64 |
| <300 | 38.1% | 33.0% | 40.6% | 40.8% | |
| 300–<500 | 36.7% | 42.0% | 32.3% | 35.7% | |
| ≥500 | 24.8% | 25.0% | 27.1% | 22.4% | |
| Rural location | 9.2% | 10.0% | 9.4% | 8.2% | 0.90 |
| Annual STEMI volume | 93 (62, 146) | 104 (71, 156) | 89 (63, 147) | 86 (62, 134) | 0.21 |
| Annual primary PCI volume | 70 (49, 105) | 75 (52, 108) | 68 (48, 104) | 66 (49, 101) | 0.38 |
Exclusions related to difficult vascular access, delays in providing consent, and other were relatively similar among the 3 groups, with absolute differences of <2% ( Table 3 ). In contrast, hospitals in the low exclusion group were more likely to exclude patients due to cardiac arrest and need for intubation, whereas hospitals in the high exclusion group were more likely to exclude patients due to difficulty in crossing the lesion ( Table 3 ).
| Variable | All Patients (n = 43,909) | Low Proportion of Non-Systems Delays (N = 13,441) | Intermediate Proportion of Non-Systems Delays (N = 18,090) | High Proportion of Non-Systems Delays (N = 12,738) |
|---|---|---|---|---|
| Difficult vascular access | 7.6% | 7.7% | 8.0% | 7.2% |
| Cardiac arrest/need for intubation | 38.4% | 42.2% | 40.5% | 35.7% |
| Delay in providing consent | 5.4% | 5.6% | 5.4% | 5.3% |
| Difficulty crossing lesion | 19.8% | 17.0% | 17.4% | 22.5% |
| Other | 28.9% | 27.6% | 28.8% | 29.4% |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
