Method
This study was approved by the Medical Research Council
(34858-3/2019/EKU). The Hungarian Myocardial Infarction Registry (HUMIR)
is a prospective and comprehensive disease registry for patients with
AMI. Since 1 January 2010, this online registry has collected personal
and clinical data on consecutive patients treated for AMI events in
Hungary, a Central European country with 9.8 million residents and a
centralised healthcare system. The registry was operated on a voluntary
basis between 2010 and 2013; however, by law, registration of all
patients with myocardial infarction in the HUMIR has been mandatory for
all healthcare providers since 1 January 2014 (statute CCXLVI/2013 of
Hungary). Data capture covered 178 structured categories, including
prehospital data, medical history, hospital treatment and coronary
interventions. The data were recorded in the online database by
healthcare professionals and administrators based on hospital
documentation, and the information was continuously monitored and
validated online. Outcome data—including vital status and repeated
hospitalisation—were regularly received from the electronic database
of the Hungarian Health Care Insurance Fund and National Central
Statistical Office.
The aim of the registry is two-fold: to provide quality assurance for
patient care and to build a database for research. All hospitals
offering acute care participate in the programme, registering
approximately 15,000 AMI cases annually; in 2021, the database contained
122,774 patients and 134,986 events. The National Health Insurance Fund
is responsible for the financing of healthcare services, whereas acute
inpatient care is funded by implementing a system similar to the
American Diagnosis Related Groups (16). The diagnosis and treatment
protocol for AMI are in line with the current European guidelines (17);
thus, the treatment of myocardial infarction with ST-elevation (STEMI)
cases is performed at centres prepared for invasive interventions. In
Hungary, the capital has six invasive centres and the remaining regions
of the country have 14. The operating principles of the centres in the
capital and country differ to a certain extent; the six centres in the
capital treat patients from the CR between 8 a.m. and 6 p.m. based on a
regional principle. Between 6 p.m. and 8 a.m., these centres care for
all patients in the region, whereas there is 24/7 care available in all
countryside centres. According to the professional protocol for
infarction care, the patient has to be transferred directly to a care
centre providing cardiac catheterisation in the case of STEMI; if this
is not possible, pharmacological anti-thrombotic therapy has to be
introduced. In daily clinical practice, pharmacological
revascularisation (thrombolysis) is rarely performed (0.1-0.3% of all
cases).
In the case of myocardial infarction without ST-elevation (NSTEMI), the
decision concerning the necessity of invasive treatment is based on the
patient’s prognosis (e.g., GRACE score). Between 1 January 2016, and 31
December 2018, a total of 42,793 AMI cases were registered in the HUMIR
database, with 18,435 being STEMI cases. During the 2-year interval
(between 1 January 2017 and 31 December 2018), the emergency care data
of 6,878 patients treated for myocardial infarction were also available;
thus, in this patient cohort, it was possible to examine the time
interval between the onset of symptoms and the reopening of the culprit
vessel (total ischaemic time, TIT). The time interval between the onset
of symptoms and the calling of emergency services was examined, along
with the different time intervals of the emergency care process: arrival
to location (T1), treatment on-site (T2) and time interval between
transferring the patient from the original location to the hospital
(T3). Time values were indicated by median time and quartiles, and we
analysed the all-cause mortality of all cases observed until 30 November
2019. The R statistical programme package (version 3.6.3;
https://www.r-project.org/about.html) was used for data
processing; in the descriptive statistical analysis, we represented the
distribution of category variables by frequency (and proportion value)
and continuous variables by mean value (distribution) and median
treatment time (quartiles). For the comparative analysis of the examined
groups, we applied the Wilcoxon test for continuous variables in
two-sample cases and the Kruskal–Wallis test in multiple-sample cases;
for categorical variables, the Chi-squared test was used (WHO No. 95).
For categorical variables, regional differences were evaluated using the
Chi-squared and accompanying post-hoc test. The calculations of
age-standardised rates in the particular regions were calculated by
applying the direct method based on the official WHO guidelines. For the
standard age group distribution, we applied the European age standard.
When analysing fatalities, we used the stratified Cox-PH regression
model (18). Because one of the major conditions of the applicability of
the Cox regression model was not met (with the help of the Schoenfeld
test, we determined that the risk ratios of the examined factors were
not time-independent), the time variable was divided into three parts
based on 30 and 365 days. Thus, this study aimed to determine the risk
ratios of individual factors.