Developing countries face challenges in providing the best reperfusion strategy for patients with ST-segment elevation myocardial infarction because of limited resources. This causes wide variation in the provision of cardiac care. The aim of this study was to assess the impact of variation in cardiac care provision and reperfusion strategies on patient outcomes in Malaysia. Data from a prospective national registry of acute coronary syndromes were used. Thirty-day all-cause mortality in 4,562 patients with ST-segment elevation myocardial infarctions was assessed by (1) cardiac care provision (specialist vs nonspecialist centers), and (2) primary reperfusion therapy (thrombolysis or primary percutaneous coronary intervention [P-PCI]). All patients were risk adjusted by Thrombolysis In Myocardial Infarction (TIMI) risk score. Thrombolytic therapy was administered to 75% of patients with ST-segment elevation myocardial infarctions (12% prehospital and 63% in-hospital fibrinolytics), 7.6% underwent P-PCI, and the remainder received conservative management. In-hospital acute reperfusion therapy was administered to 68% and 73% of patients at specialist and nonspecialist cardiac care facilities, respectively. Timely reperfusion was low, at 24% versus 31%, respectively, for in-hospital fibrinolysis and 28% for P-PCI. Specialist centers had statistically significantly higher use of evidence-based treatments. The adjusted 30-day mortality rates for in-hospital fibrinolytics and P-PCI were 7% (95% confidence interval 5% to 9%) and 7% (95% confidence interval 3% to 11%), respectively (p = 0.75). In conclusion, variation in cardiac care provision and reperfusion strategy did not adversely affect patient outcomes. However, to further improve cardiac care, increased use of evidence-based resources, improvement in the quality of P-PCI care, and reduction in door-to-reperfusion times should be achieved.
In developing countries, there is wide variation in the provision of cardiac care, and there are challenges in providing the best reperfusion strategy as recommended by international guidelines for ST-segment elevation myocardial infarction (STEMI). The recommended reperfusion strategy for STEMI is the percutaneous coronary intervention. Nevertheless, as in other developing countries, the primary reperfusion strategy in Malaysia is thrombolytic therapy. Malaysia has a dual public and private health care system, with 69% of the country’s hospitalizations occurring in the public system. The challenges faced in cardiac care provision are limited resources and a relative lack of specialist cardiac care centers with primary percutaneous coronary intervention (P-PCI) facilities for STEMI management. At present, publicly funded cardiac care facilities equipped for P-PCI exist mainly in the larger, more economically developed urban areas. The aim of this study was to assess the impact of variation in cardiac care provision and reperfusion strategies on outcomes of patients with STEMIs.
Methods
The Malaysian National Cardiovascular Disease Database–Acute Coronary Syndrome Registry (NCVD-ACS) commenced in January 2006 and is a prospective registry of patients who present with acute coronary syndromes (ACS). Patient recruitment occurs at 16 hospitals with varying cardiac facilities, 15 of which are publicly funded (14 from the Ministry of Health, 1 university hospital), and the corporatized National Heart Institute. The 15 public hospitals represent 71% of all publicly funded hospitals with cardiac care services in the country, including the main tertiary referral center of Malaysia. Of these 16 participating sites, 6 provide specialist cardiac care services and have on-site P-PCI and cardiac surgical services. These 6 centers perform in total approximately 3,400 to 3,600 PCI procedures per year. One center performs >2,000 procedures per year, and 4 others perform 300 to 400 procedures per year. At all 6 centers, 90% are elective PCI procedures. The remaining 10 sites are nonspecialist cardiac care facilities. Their cardiac care services are run by physicians and do not have on-site P-PCI or cardiac surgical backup. These differences in cardiac care provision are due to financial and human resource constraints faced by the public health care service. There are no private, for-profit centers reporting to the NCVD-ACS.
Details of medical history, presenting symptoms, reperfusion therapy, in-hospital clinical care, and health outcomes through 1 year after ACS are recorded for all patients aged ≥18 years. Ethical approval for the NCVD-ACS was obtained from the Ministry of Health Medical Research and Ethics Committee. A waiver of the requirement for informed consent was approved by the Medical Research and Ethics Committee. As an alternative, a public notice is displayed at all sites, and patients are given the option to opt out of the NCVD-ACS. Data from anonymized patients registered from January 1, 2006 through December 31, 2008 with 30-day follow-up was used. For this study, all patients presenting with STEMIs were selected.
The diagnosis and confirmation of STEMI were based on signs and symptoms of ACS (chest pain or overwhelming shortness of breath), elevated serum cardiac biomarkers, and ST-segment elevation in contiguous leads on electrocardiography or the development of new left bundle branch block. Clinical care provided was at the discretion of the treating physician or cardiologist. A medical history stated to be “not known” or “not recorded” in the registry records was classified as absent for the purposes of analysis in this study.
Risk stratification by severity of STEMI presentation was done using the Thrombolysis In Myocardial Infarction (TIMI) risk score for STEMI, which has been validated in the Malaysian population. The TIMI risk score consists of the following components: age; systolic blood pressure; heart rate; Killip classification; infarct location or left bundle branch block; history of diabetes, hypertension, or angina pectoris; weight; and time to reperfusion (thrombolysis or P-PCI). Time to reperfusion was modified for the Malaysian population to be door-to-needle time or door-to-balloon time instead of symptom-to-reperfusion time because of insufficient information on symptom onset to presentation time. Patients who did not undergo any reperfusion strategy were given a score of 1 for the component of time to reperfusion (representing the worst score for that variable). The TIMI STEMI scoring mechanism has been published and is given in the Appendix . Scores of ≤3, 4 to 6, and ≥7 are in the low-, intermediate-, and high-risk categories, respectively.
The clinical outcome of interest was 30-day all-cause mortality, which included in-hospital mortality. Mortality details were obtained through hospital records and phone calls to patients or relatives after 30 days. Annual confirmation of mortality for the NCVD-ACS is done through record linkages with the Malaysian National Registration Department for deaths. IBM InfoSphere QualityStage ( http://www-01.ibm.com/software/data/infosphere/qualitystage/ ) was used for record-matching purposes. Rule sets for record matching were prepared on the basis of methods available in the software. These were executed for key identifier variables such as name, identification card number, and year and month of birth. All Malaysians have unique numeric identification numbers, thereby enabling accurate record linkages.
Outcomes comparisons were analyzed for cardiac care provision and acute reperfusion therapies: (1) cardiac care provision (specialist cardiac care vs nonspecialist cardiac care facilities), and (2) acute reperfusion therapies (thrombolysis [pre- and in-hospital], P-PCI, and conservative clinical management).
Data analysis that ignores missing data (complete case analysis) has been shown to produce biased results, so missing data were imputed. Missing data were reviewed to determine if they were missing at random before imputation. Five variables with missing values <5% had mean or median values imputed: age 0.1%, systolic blood pressure 1.0%, heart rate at presentation 1.9%, gender 2.1%, and smoking status 4.1%. Two variables, door-to-reperfusion time (21.6%) and weight (36.0%), were imputed using single imputation with a random error term method. For sensitivity analyses, we ran a complete case analysis, which did not materially change the results (direction and statistical significance) of this study.
Baseline differences between patients presenting at facilities with and without specialist cardiac care services were determined using unequal-variance Student’s t tests for continuous variables and Pearson’s chi-square tests for categorical variables. P values for differences in the severity of STEMI presentation by cardiac care provision and reperfusion therapies were tested using Wilcoxon Mann-Whitney tests. Comparisons of acute reperfusion therapies and crude mortality rates by cardiac care provision were tested using Pearson’s chi-square tests. Comparisons of in-hospital treatment and adjusted mortality rates were assessed using generalized estimation equations to adjust for clustering effects of cardiac care facilities. An independent working model structure was used. Adjusted mortality rates were obtained using the estimated marginal means, and differences were tested using Wald chi-square tests. Comparisons of overall and in-hospital fibrinolysis 30-day mortality rates for cardiac care facilities were adjusted for STEMI severity (described by TIMI risk score). Center 30-day mortality rates for prehospital fibrinolysis, conservative therapy, and reperfusion strategies for specialist cardiac care facilities were adjusted for STEMI severity as a linear term because of the small number of events. Two-sided p values <0.05 were considered statistically significant. All analyses were performed using SPSS version 17.0 for Windows (SPSS, Inc., Chicago, Illinois).
Results
There were 10,682 patients registered in the NCVD-ACS from January 1, 2006 through December 31, 2008, 4,701 with STEMIs. One hundred thirty-nine patients (3%) were excluded because of missing reperfusion therapy data. Of the 4,562 patients with STEMIs studied, 75% were given fibrinolytics (12% prehospital and 63% in-hospital), 7.6% underwent P-PCI, and the remainder received conservative management. Streptokinase accounted for up to 98% of fibrinolytics given in-hospital.
The racial distribution was not representative of the Malaysian population, with a higher proportion of Indians than the general population and lower proportions of other races ( Table 1 ). Those presenting to specialist cardiac care facilities had higher proportions of hypertension, diabetes, renal impairment, coronary artery disease with angiographically documented stenoses of >50%, and previous myocardial infarctions, as well as higher β-blocker and lipid-lowering drug use. Despite higher proportions of risk factors and previous coronary artery disease in patients with STEMIs presenting to specialist cardiac care facilities, there was no statistical difference in the severity of STEMI presentation, as described by the TIMI risk score.
| Variable | Overall Population (n = 4,562) | Specialist Cardiac Care Facilities (n = 2,501) | Nonspecialist Cardiac Care Facilities (n = 2,061) | p Value ∗ |
|---|---|---|---|---|
| Age (yrs) | 56 (48–65) | 56 (48–65) | 56 (48–65) | 0.32 |
| Women | 15% | 15% | 15% | |
| Weight (kg) | 67 (58–77) | 67 (59–76) | 68 (57–78) | 0.12 |
| Malay | 53% | 44% | 64% | |
| Chinese | 20% | 27% | 12% | |
| Indian | 18% | 21% | 14% | |
| Others | 8.4% | 8% | 9% | |
| Smoking status | 0.001 | |||
| Current | 51% | 51% | 51% | |
| Past | 20% | 18% | 22% | |
| Never | 29% | 31% | 27% | |
| Diabetes mellitus | 37% | 38% | 34% | 0.003 |
| History of hypertension | 49% | 51% | 46% | 0.001 |
| Renal impairment (moderate to severe) | 3.2% | 4% | 2.2% | 0.001 |
| Previous myocardial infarction | 9.5% | 11.6% | 6.9% | 0.001 |
| Peripheral vascular disease | 0.3% | 0.3% | 0.3% | 0.86 |
| Cerebrovascular disease | 2.7% | 3% | 2.5% | 0.32 |
| Previous angina pectoris | 52% | 52% | 53% | 0.52 |
| Documented coronary artery disease (>50% stenosis) | 5.6% | 8.4% | 2.3% | 0.001 |
| Diabetes/hypertension/previous angina pectoris | 80% | 80% | 80% | 0.94 |
| Medications at presentation | ||||
| β blockers | 13% | 16.2% | 9% | 0.001 |
| Calcium channel blockers | 5.1% | 5.8% | 4.2% | 0.02 |
| Lipid-lowering agents | 17% | 20% | 13% | <0.001 |
| Antiarrhythmic agents | 1.8% | 2% | 1.7% | 0.51 |
| Presenting characteristics | ||||
| Infarct location | ||||
| Anterior or left bundle branch block | 60% | 59% | 61% | 0.23 |
| Inferior | 46% | 46% | 46% | 0.89 |
| Killip class II–IV | 29% | 31% | 28% | 0.04 |
| Heart rate >100 beats/min | 18% | 17% | 18% | 0.56 |
| Systolic blood pressure <100 mm Hg | 8.4% | 8.6% | 8.1% | 0.58 |
| TIMI risk score | 4 (2–5) | 4 (2–5) | 4 (2–5) | 0.66 |
| TIMI risk score categories | 0.90 | |||
| Low (score ≤3) | 48% | 48% | 48% | |
| Intermediate (score 4–6) | 36% | 36% | 37% | |
| High (score ≥7) | 16% | 16% | 16% |
Specialist cardiac care facilities received a 5% higher proportion of referred patients treated with prehospital fibrinolytics than nonspecialist cardiac care facilities ( Table 2 ). Timely reperfusion of <30 minutes of door-to-needle time for fibrinolytic therapy and <90 minutes of door-to-balloon time for P-PCI was low, about 1 in 4. Specialist cardiac care facilities had a significantly lower proportion of timely reperfusion for fibrinolytic therapy compared with nonspecialist facilities. There were significant differences in in-hospital drug prescription by type of cardiac care facility. There was consistently higher use of aspirin, adenosine diphosphate antagonists, β blockers, statins, and unfractionated heparin at the specialist cardiac care facilities. However, there was lower use of low–molecular weight heparin. With increasing severity of STEMI presentation, the proportion of patients receiving P-PCI increased from 2.8% for TIMI risk score 0 to 33% for TIMI risk score >8 (p for trends <0.001; Figure 1 ).
| Variable | Specialist Cardiac Care Facilities (n = 2,501) | Nonspecialist Cardiac Care Facilities (n = 2,061) | p Value |
|---|---|---|---|
| Intervention | |||
| Fibrinolytics given in-hospital | 1,363 (55%) | 1,509 (73%) | <0.001 |
| Prehospital fibrinolytics given | 359 (14%) | 190 (9%) | <0.001 |
| Primary PCI | 346 (14%) | NA | |
| No emergency reperfusion | 433 (17%) | 359 (17%) | 0.93 |
| Door-to-reperfusion >4 h | 891 (36%) | 716 (35%) | 0.53 |
| Door-to-needle time <30 minutes ∗ | 327 (24%) | 469 (31%) | <0.001 |
| Door-to-balloon time <90 minutes | 95 (28%) | NA | |
| In-hospital medications | |||
| Aspirin | 2,408 (96%) | 1,870 (91%) | 0.001 |
| Adenosine diphosphate antagonists | 2,127 (85%) | 1,004 (49%) | 0.03 |
| Glycoprotein receptor inhibitors | 134 (5.4%) | 27 (1.3%) | 0.002 |
| Unfractionated heparin † | 382 (15%) | 88 (4.3%) | 0.006 |
| Low–molecular weight heparin † | 683 (27%) | 889 (43%) | 0.08 |
| Statins | 2,345 (94%) | 1,780 (86%) | <0.001 |
| Other lipid-lowering agents | 75 (3%) | 116 (5.6%) | 0.04 |
| β blockers | 1,782 (71%) | 1,159 (56%) | 0.001 |
| Calcium channel blockers | 204 (8.2%) | 112 (5.4%) | 0.07 |
| Diuretics | 641 (26%) | 525 (26%) | 0.97 |
| Angiotensin-converting enzyme inhibitors | 1,413 (57%) | 1,202 (58%) | 0.81 |
| Angiotensin II receptor blockers | 137 (5.5%) | 81 (3.9%) | 0.35 |
| Antiarrhythmic agents | 189 (7.6%) | 168 (8.2%) | 0.77 |
| 30-day mortality rates | |||
| Prehospital fibrinolytics given | |||
| Crude mortality rate | 12.5% (45) | 8.4% (16) | 0.83 |
| Adjusted mortality rate (95% CI) ‡ | 9% (7%–11%) | 7% (4%–9%) | 0.18 |
| Fibrinolytics given in-hospital | |||
| Crude mortality rate | 9.5% (130) | 8.9% (134) | 0.78 |
| Adjusted mortality rate (95% CI) ‡ | 10% (7%–14%) | 9% (4%–14%) | 0.68 |
| Primary PCI | |||
| Crude mortality rate | 11.8% (41) | NA | |
| Adjusted mortality rate (95% CI) ‡ | 7% (3%–11%) | NA | |
| No emergency reperfusion | |||
| Crude mortality rate | 17.1% (74) | 16.2% (58) | 0.66 |
| Adjusted mortality rate (95% CI) ‡ | 12% (10%–15%) | 11% (3%–19%) | 0.77 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
