Only a few studies have focused on the clinical characteristics and outcomes of non-ST-segment myocardial infarction (NSTEMI) during off-hours. The purpose of this study was to compare the impact of patients’ arrival time on the care of NSTEMI and whether this pattern might affect hospital mortality. This study analyzed 4,736 NSTEMI patients included in the Korea Acute Myocardial Infarction Registry from November 2005 to January 2008. Patients’ arrival time was classified into regular hours (weekdays, 9:00 a.m. to 6:00 p.m. ) and off-hours (weekdays 18:01 p.m. to 8:59 a.m. , weekends, and holidays). A subtotal of 2,225 (46.9%) patients was admitted during off hours, compared with 2,511 (53.1%) patients with regular-hour admission. A higher proportion of patients admitted during off-hours had a higher Killip class, had more frequent cardiopulmonary resuscitation, were less likely to receive percutaneous coronary intervention (PCI) (67.7% vs 72.7%, p <0.001), and had longer door-to-balloon times (28 hours, interquartile range: 11 to 63 vs 23 hours, interquartile range 4 to 67, p <0.001). Although unadjusted hospital mortality was associated with admission during off-hours (4.5% vs 3.3%, p = 0.023), after adjusting for all patients covariates, the difference in mortality was attenuated and was no longer statistically significant (odds ratio 0.94, 95% confidence interval 0.59 to 1.48, p = 0.793). In conclusion, despite receiving fewer PCIs and having substantially longer waiting times to PCI, patients admitted during off-hours may not be at risk for increased in-hospital mortality. If patients are treated within an appropriate reperfusion strategy according to their clinical risk, arrival time may not influence on mortality.
Several recent studies have found that patients admitted during off-hours (weekday nights, weekends, and holidays) were less likely to receive timely reperfusion after acute myocardial infarction. For ST-segment-elevation myocardial infarction (STEMI), conflicting results exist on the outcome during off-hours. Many factors, including nurse and medical staffing, medical level of education, logistics (number of beds), severity of disease, and co-morbidities, were associated with mortality variations. However, only a few studies have focused on the clinical characteristics and outcomes of NSTEMI during off-hours. The purpose of this study was to compare the impact of patients’ arrival time (regular hours vs off-hours) on clinical characteristics in patients with NSTEMI and whether this pattern might affect the hospital mortality in clinical practice.
Methods
The study population, enrolled in a nationwide prospective Korea Acute Myocardial Infarction Registry (KAMIR) between November 2005 and January 2008, consisted of 4,736 patients with NSTEMI (3,116 men, 62.0 ± 12.8 years).
The KAMIR, launched in November 2005, was a Korean prospective multicenter data collection registry reflecting real-world treatment practices and outcomes in Asian patients diagnosed with acute myocardial infarction. The registry includes data from 50 hospitals with facilities for primary percutaneous coronary (PCI) and on-site cardiac surgery. Data were collected by a trained study coordinator using a standardized case report form and protocol. All management decisions were at the discretion of the treating cardiologists. Before the initiation of the KAMIR study, several investigator meetings were held, and a steering committee was selected to standardize the care given in clinical practice to minimize the differences in medical care among the hospitals. The ethics committee at each participating institution approved the study.
Eligible patients were ≥18 years of age at the time of hospital presentation, had to be admitted for an NSTEMI, and had ≥1 of the following: electrocardiographic changes consistent with an NSTEMI (defined as ST-segment depression and T-wave inversion of 0.2 mV in 2 contiguous leads), serial increases in serum biochemical markers of myocardial necrosis, and documentation of coronary artery disease by coronary angiogram. Patients with ST-segment elevation or presumed new left bundle branch block were excluded, and ST-segment depression being defined as new horizontal or downsloping depression 0.05 mV in 2 contiguous leads. Cutoff value of cardiac troponin was defined as the lowest concentration with a value of coefficient of variation (CV) <10%. Patients’ arrival time was classified into regular hours (weekdays, 9:00 a.m. to 6:00 p.m. ) and off-hours (weekdays 18:01 p.m. to 8:59 a.m. , weekends, and holidays). The patients with NSTEMI who underwent coronary angiography and PCI within 24 hours after arrival were classified into the early invasive intervention. Door-to-balloon time was measured as the time from hospital arrival to the time of any therapeutic device (balloon, stent, or thrombectomy catheter) in the infarct-related artery. Global Registry of Acute Coronary Events (GRACE) risk scores were calculated in all patients. Age, initial vital signs (systolic blood pressure and heart rate), and Killip classification on admission, ST-segment depression on electrocardiogram, and blood chemistry (baseline serum creatinine) were implemented. GRACE risk scores were classified as low-, intermediate-, and high-risk groups ( www.outcomes-umassmed.org/grace ; low risk, 1 to 108; intermediate risk, 109 to 140; and high risk, 141 to 372 for in-hospital mortality). Baseline characteristic and hospital outcomes were analyzed in the overall NSTEMI cohort and in the low- to intermediate-risk and high-risk subpopulations. Primary end point was in-hospital death from any cause after onset of myocardial infarction.
Continuous data are expressed as the mean ± SD or the median and interquartile range (twenty-fifth and seventy-fifth percentiles). Categorical data are expressed as percentages. Continuous variables were analyzed using Student’s t test or the Mann-Whitney U test, as appropriate, and categorical variables were compared using the chi-square test. The Breslow-Day test was performed to assess the homogeneity of the relative risk for hospital mortality across participating centers. Univariate and multivariate analysis were performed to estimate the prognostic significance of clinical variables for in-hospital mortality. All available variables considered potentially relevant were included: age, gender, cardiopulmonary resuscitation on admission, Killip class on admission, primary ventricular tachyarrhythmia, cardiovascular risk factors (hypertension, dyslipidemia, smoking, diabetes mellitus, family history of coronary heart disease), previous myocardial infarction, chronic heart failure, and previous cerebrovascular disease, peripheral arterial disease, chronic lung disease, admission during off-hours, PCI performed, cardiogenic shock, renal function (creatinine >1.5), left ventricular ejection fraction <40%, and GRACE risk score. Variables with p values <0.1 on univariate model were entered into multivariate logistic regression models. We used stepwise elimination and backward selections to select the most powerful predictive variables. p value <0.05 was considered statistically significant. Analyses were performed using the Statistical Package for Social Sciences, version 18.0 (SPSS, Chicago, IL, USA).
Results
As shown in Figure 1 , 5,082 patients with NSTEMI were enrolled in KAMIR. From this population, 346 (6.8%) patients were excluded because of missing or invalid arrival dates or times. Therefore, the study population consisted of 4,736 patients: 2,225 (46.9%) patients were admitted during off-hours and 2,511 (53.1%) patients during regular hours. Table 1 lists baseline characteristics according to patients’ arrival time and clinical risk defined by GRACE score. Baseline characteristics were similar. Compared with patients admitted during regular hours, however, a significantly higher proportion of patients admitted during off-hours had Killip class III and IV, received cardiopulmonary resuscitation on presentation, were less likely to receive PCI (67.7% vs 72.7%, p <0.001), and had longer door-to-balloon times (28 hours 1 min, interquartile range [IQR]: 11:27 to 63:04 vs 23 hours 49 min, IQR: 4:50 to 67:03 min, p <0.001). Among patients who underwent reperfusion therapy, the proportion of patients achieving early invasive PCI was much lower during off-hours (45.1% vs 50.6%, p = 0.002). Cardiac enzymes including creatine kinase-MB, troponin I, creatinine >1.5 mg/dl, pro–brain natriuretic peptide, and glucose were significantly higher in patients admitted during off-hours. Patients at high risk (GRACE risk score >140) were admitted more often during off-hours. In subgroup analysis, however, the proportion of patients who underwent early invasive PCI (44.3% vs 45.9%, p = 0.341) and time to PCI (34 hours 25 min, IQR: 10:58 to 84:31 vs 27 hours, IQR: 4:50 to 89:06, p = 0.342) were not different between regular hours and off-hours in patients at high risk.
| Variable | Overall | High Risk | Low to Intermediate Risk | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Regular Hours (n = 2,511) | Off-Hours (n = 2,225) | p Value | Regular Hours (n = 1,114) | Off-Hours (n = 1,050) | p Value | Regular Hours (n = 1,397) | Off-Hours (n = 1,175) | p Value | |
| Age (yrs) | 64.41 ± 12.15 | 64.4 ± 12.66 | 0.98 | 71.85 ± 9.09 | 72.48 ± 9.06 | 0.10 | 58.47 ± 10.95 | 57.18 ± 10.94 | 0.003 |
| Men | 1,646 (65.7) | 1,447 (65.1) | 0.70 | 483 (43.4) | 492 (46.9) | 0.10 | 378 (27.1) | 283 (24.1) | 0.08 |
| Hypertension | 1,372 (55.1) | 1,177 (53.4) | 0.26 | 654 (59.4) | 617 (59.4) | 0.98 | 718 (51.6) | 560 (48.1) | 0.07 |
| Diabetes mellitus | 788 (31.8) | 740 (33.6) | 0.18 | 429 (39.0) | 422 (40.7) | 0.42 | 359 (26.0) | 318 (27.2) | 0.47 |
| Hyperlipidemia | 296 (13.4) | 288 (14.9) | 0.18 | 109 (11.4) | 119 (13.1) | 0.24 | 187 (15.0) | 169 (16.4) | 0.38 |
| Smoker | 1,304 (52.3) | 1,155 (52.3) | 0.99 | 472 (42.7) | 425 (41.0) | 0.41 | 832 (60.0) | 730 (62.4) | 0.22 |
| Systolic blood pressure (mm Hg) | 131.07 ± 27.06 | 134.46 ± 29.77 | 0.001 | 122.38 ± 27.69 | 127.87 ± 31.29 | 0.001 | 137.99 ± 24.43 | 140.34 ± 27.03 | 0.021 |
| Diastolic blood pressure (mm Hg) | 78.46 ± 16.22 | 80.26 ± 17.10 | 0.001 | 73.38 ± 16.39 | 76.24 ± 17.63 | 0.001 | 82.52 ± 14.90 | 83.84 ± 15.78 | 0.029 |
| Heart rate (beats/min) | 78.11 ± 18.95 | 80.81 ± 22.06 | 0.01 | 82.75 ± 22.81 | 87.48 ± 25.29 | 0.001 | 74.41 ± 14.13 | 74.85 ± 16.55 | 0.47 |
| Killip class I | 1,901 (75.7) | 1,558 (70.0) | 0.001 | 584 (52.4) | 478 (45.5) | 0.001 | 1,317 (94.3) | 1,080 (91.9) | 0.018 |
| Killip class II | 323 (12.9) | 301 (13.5) | 0.50 | 250 (22.4) | 221 (21.0) | 0.43 | 73 (5.2) | 80 (6.8) | 0.09 |
| Killip class III | 241 (9.6) | 285 (12.8) | 0.001 | 234 (21.0) | 270 (25.7) | 0.01 | 7 (0.5) | 15 (1.3) | 0.03 |
| Killip class IV | 46 (1.8) | 81 (3.6) | 0.001 | 46 (4.1) | 81 (7.7) | 0.001 | 0 (0) | 0 (0) | 1.00 |
| CPR on presentation | 22 (0.9) | 36 (1.6) | 0.02 | 22 (2.0) | 35 (3.3) | 0.049 | 0 (0) | 1 (0.1) | 0.27 |
| VT/VF on presentation | 41 (1.6) | 38 (1.7) | 0.83 | 36 (3.2) | 28 (2.7) | 0.44 | 5 (0.4) | 10 (0.9) | 0.10 |
| PCI | |||||||||
| Early PCI (<24 h) | 924 (50.6) | 679 (45.1) | 0.002 | 318 (45.9) | 263 (44.3) | 0.34 | 606 (53.5) | 416 (46.3) | 0.001 |
| Delayed PCI (>24 h) | 902 (49.3) | 827 (54.9) | 375 (54.1) | 345 (56.7) | 527 (46.5) | 482 (53.7) | |||
| Door-to-balloon time (h:min) ∗ | 23:49 (4:50, 67:03) | 28:01 (11:27, 63:04) | 0.009 | 27:00 (4:50, 89:06) | 34:25 (10:58, 84:31) | 0.34 | 22:44 (4:51, 51:12) | 26:29 (11:43, 55:51) | 0.009 |
| No reperfusion | 685 (27.3) | 719 (32.3) | 0.001 | 421 (37.8) | 442 (42.1) | 0.04 | 264 (18.9) | 277 (23.6) | 0.004 |
| CK-MB (peak) | 66.27 ± 211.822 | 83.07 ± 237.81 | 0.01 | 66.87 ± 120.55 | 92.82 ± 328.21 | 0.01 | 65.79 ± 262.77 | 74.23 ± 102.59 | 0.30 |
| Troponin I (peak) | 20.77 ± 51.51 | 23.55 ± 44.31 | 0.07 | 22.91 ± 53.04 | 27.19 ± 55.49 | 0.09 | 19.01 ± 50.17 | 20.21 ± 30.32 | 0.52 |
| Creatinine (>1.5) | 306 (12.2) | 341 (15.3) | 0.002 | 243 (21.8) | 290 (27.6) | 0.002 | 63 (4.5) | 51 (4.3) | 0.82 |
| pro-BNP | 3,138 ± 6,592 | 3,788 ± 7,629 | 0.016 | 5,763 ± 8,646 | 5,800 ± 9,762 | 0.049 | 1,277 ± 3,599 | 1,060 ± 3,005 | 0.20 |
| Glucose | 156.88 ± 79.30 | 174.71 ± 91.86 | 0.001 | 169.21 ± 88.79 | 196.79 ± 109.38 | 0.001 | 147.03 ± 69.28 | 154.94 ± 66.80 | 0.004 |
| LDL mg/dl | 116.42 ± 42.64 | 116.85 ± 51.57 | 0.76 | 109.63 ± 45.81 | 112.91 ± 57.44 | 0.17 | 121.61 ± 39.28 | 120.13 ± 45.85 | 0.40 |
| Ejection fraction (%) | 53.2 ± 12.58 | 52.07 ± 13.38 | 0.005 | 48.91 ± 13.46 | 47.45 ± 14.21 | 0.002 | 56.58 ± 10.70 | 56.15 ± 11.11 | 0.33 |
| Ejection fraction (<40%) | 362 (15.7) | 421 (20.8) | 0.001 | 272 (26.8) | 322 (33.9) | 0.001 | 90 (7.0) | 99 (9.2) | 0.049 |
| GRACE risk score | |||||||||
| Low (≤108) | 583 (23.2) | 532 (23.9) | 0.57 | ||||||
| Intermediate (109–140) | 814 (32.4) | 643 (28.9) | 0.009 | ||||||
| High (>140) | 1,114 (44.4) | 1,050 (47.2) | 0.051 | ||||||
Figure 2 shows reperfusion strategies according to patients’ arrival time and risk. Patients admitted during off-hours more often did not undergo reperfusion compared with those with regular-hours admission (32.3% vs 27.3%, p <0.001). In the subgroup analysis, patients at high risk more often did not undergo reperfusion compared with those who were at low risk in our study (40.2% vs 21.1%, p <0.001).
Figure 3 shows distribution of the proportion of patients who underwent early invasive PCI (within 24 hours) after admission according to patients’ arrival period and clinical risk. Patients admitted during regular hours received PCI during the day of admission or the following day (M-shaped pattern). Off-hours patients received PCI immediately or waited until day. These 2 patterns were similar in patients at high and low risk ( Figure 3 ).
In-hospital complications and mortality are summarized in Table 2 . Test results for assessing the homogeneity of the relative risk across centers were not significant for hospital mortality (Breslow-Day test, p = 0.215). The mortality rate was significantly higher for the patients admitting during off-hours compared with those admitted during regular hours throughout the index hospitalization (4.5% vs 3.3%, p = 0.023) (odds ratio [OR]: 1.409, 95% confidence interval [CI] : 1.04 to 1.89, p value = 0.024) Overall, after adjusting for all patients covariates, the difference in mortality between the regular-hours and off-hours periods was attenuated and was no longer statistically significant (OR: 0.94, 95% CI: 0.59 to 1.48, p = 0.793).
| Overall | High Risk | Low to Intermediate Risk | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Regular Hours (n = 2,511) | Off-Hours (n = 2,225) | p Value | Regular Hours (n = 1,114) | Off-Hours (n = 1,050) | p Value | Regular Hours (n = 1,397) | Off-Hours (n = 1,175) | p Value | |
| Hospital complication | |||||||||
| Heart failure | 20 (0.8%) | 27 (1.2%) | 0.148 | 18 (1.6%) | 26 (2.5%) | 0.156 | 2 (0.1%) | 1 (0.1%) | 0.667 |
| Acute kidney injury | 16 (0.6%) | 31 (1.4%) | 0.009 | 13 (1.2%) | 28 (2.7%) | 0.010 | 3 (0.2%) | 3 (0.3%) | 0.829 |
| Cardiogenic shock | 65 (2.6%) | 85 (3.8%) | 0.015 | 58 (5.2%) | 80 (7.7%) | 0.021 | 7 (0.5%) | 5 (0.4%) | 0.784 |
| Cerebrovascular accident | 10 (0.4%) | 14 (0.6%) | 0.261 | 5 (0.5%) | 13 (1.2%) | 0.043 | 5 (0.4%) | 1 (0.1%) | 0.154 |
| Ventilator therapy | 66 (2.7%) | 111 (5.1%) | 0.001 | 64 (5.8%) | 109 (10.6%) | 0.001 | 2 (0.1%) | 2 (0.2%) | 0.862 |
| In-hospital death | 82 (3.3%) | 101 (4.5%) | 0.023 | 77 (6.9%) | 94 (9.0%) | 0.078 | 5 (0.4%) | 7 (0.6%) | 0.379 |
| Early PCI (<24 h) | 16 (0.6%) | 17 (0.8%) | 15 (1.4%) | 16 (1.5%) | 1 (0.1%) | 1 (0.1%) | |||
| Delayed PCI (>24 h) | 15 (0.6%) | 20 (0.9%) | 13 (1.1%) | 20 (1.9%) | 2 (0.1%) | 0 (0.0%) | |||
| No reperfusion | 51 (2.0%) | 64 (2.9%) | 49 (4.4%) | 58 (5.5%) | 2 (0.1%) | 6 (0.5%) | |||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
