Myocardial infarction patients referred to the primary care physician after 1‑year treatment according to a guideline-based protocol have a good prognosis

All patients treated with a pPCI for STEMI in the LUMC are included in the prospective MISSION! MI registry [13]. For this current observational retrospective analysis we evaluated all consecutive patients treated between February 2004 up to May 2013 who, after completion of the 1‑year MISSION! MI follow-up were returned to the care of their GP. Patients who died during the first year after their index infarction, or patients who were transferred during admission to another hospital due to logistic reasons were not included in this analysis. Logistic reasons were lack of available space for patients to admit, patient’s preference or when patients were transferred back to the referring hospital after the pPCI. STEMI was defined as typical electrocardiographic changes (ST-elevation ≥0.2 mV in ≥2 contiguous leads in V₁ through V₃, ≥0.1 mV in other leads, or presumed new left bundle branch block) and a typical rise and fall of cardiac biomarkers accompanied by chest pain for at least 30 min [14]. Since the data did not contain any identifiers that could be traced back to the individual patient and was obtained for patient care, the Dutch Central Committee on Research involving Human Subjects (CCMO) permits the use of anonymous data without prior approval of an institutional review board. This study was conducted according to the declaration of Helsinki.

The institutional, guideline-based MISSION! MI protocol is a standardised clinical framework which consists of a pre-hospital, in-hospital and an outpatient phase to optimise clinical decision making and treatment up to 1 year after the index event [13, 15, 16]. The MISSION! MI protocol is in accordance with the current STEMI guidelines and was adjusted when required [15, 16]. In the pre-hospital phase, a high-quality 12-lead electrocardiogram was obtained. If a STEMI was diagnosed, patients were treated by the paramedics with a loading dose of clopidogrel or prasugrel, aspirin, heparin and intravenous glycoprotein IIb/IIIa inhibitors, if appropriate. During the in-hospital phase patients were directly transferred to the catheterisation laboratory for pPCI according to the current guidelines. If no contraindications existed, β‑blockers, angiotensin-converting enzyme (ACE) inhibitors and statins were administered within 24 h of admission. Dual antiplatelet therapy was additionally prescribed, consisting of aspirin 100 mg daily for life and prasugrel 10 mg daily or clopidogrel 75 mg daily for 12 months, if appropriate. During the outpatient clinic phase, 4 clinic visits were scheduled and patients were treated in accordance with current guidelines to reach the secondary prevention targets. Furthermore, several functional tests, such as a stress echocardiography and Holter registration, were obtained and, if necessary, an intervention was performed. An important part of the outpatient clinic care was emphasis on the need for drug compliance, as well as the education about, and modification of, lifestyle behaviour (smoking cessation, healthy diet, exercise and weight management). Patients also participated in a professional cardiac rehabilitation programme as part of the routine care, in which they had a dietician, psychologist and social worker at their disposal, as this has been associated with better one-year outcome [13, 17]. After 1 year of intensive monitoring, patients were, by protocol, returned to the care of their GP if they were asymptomatic with a left ventricular ejection fraction (LVEF) >45%.

In accordance with to our protocol, all risk factors, clinical features and laboratory measurements were systematically collected for each individual MISSION! patient in EPD-VISION, using a unique study code. Echocardiographic images were attained from patients at rest in left lateral decubitus position using a commercially available system (Vivid 7 and E9, GE, Healthcare, Horten, Norway). Standard M‑mode and 2D (colour, pulsed and continuous wave Doppler) images were obtained from the parasternal (long- and short-axis) and apical views (long-axis, 2‑ and 4‑chamber), using 3.5-MHz or M5S transducers, and digitally stored for offline analysis (EchoPac BT13, GE Medical Systems, Horten, Norway). The LVEF, wall motion score index (WMSI) and the grade of mitral regurgitation (MR) were measured according to the current echocardiographic recommendations [18, 19]. Clinical follow-up data were prospectively collected in the electronical patient files by independent clinicians. Data from patients were gathered from either out-patient chart reviews or by telephone interview. Information on the vital status was obtained from the Dutch municipal personal records database. Cause of death was retrieved from the GP.

The primary endpoint of this study is all-cause mortality. The secondary endpoint is a combined endpoint of coronary revascularisation, recurrent MI, implantation of an implantable cardioverter defibrillator (ICD) or a pacemaker, hospitalisation due to heart failure, stroke and death. All these adverse events combined have been defined as major adverse cardiac events (MACE).

Data are summarised as means with standard deviation in case of normally distributed or as median with interquartile ranges (IQR) in case of non-normally distributed data. Categorised data are shown as numbers with percentages. Univariable Cox proportional hazard regression models were used to assess the association of age, gender and pre-specified covariates, which are known associated variables in literature, with all-cause mortality or occurrence of time-dependent adverse events (MACE) in STEMI patients [1‐3, 20‐23]. Age, gender and other variables significant at p < 0.10 were entered into a multivariable Cox model to calibrate a combined prognostic index to predict either all-cause mortality or MACE [24]. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated. To classify GP patients into either high- or low-risk groups based on these Cox regression models, we dichotomised the prognostic index using the median value. Stratified by these two groups, Kaplan-Meier curves were then used to estimate and verify survival expectations (time to either all-cause death or MACE). The log-rank test was calculated to compare the cumulative incidences of the endpoints between the 2 groups. All statistical tests were two-tailed, p-values <0.05 were considered statistically significant. Analyses were performed with SPSS 23.0 statistical analysis software (IBM, Armonk, NY, USA).

Patients characteristics, medication use and laboratory results after 1 year MISSION! MI follow-up are summarised in Tab. 1. Mean age was 61.6 ± 11.7 year and 686 (74.4%) was male gender. In 70 (7.6%) patients LVEF was below 45%.

In total, 48 patients deceased after 1 year MISSION! MI follow-up. The cause of death was adjudicated as cardiac origin in 6 patients, likely cardiac in 3 patients, non-cardiac in 35 patients, unlikely cardiac in 2 patients and the cause of death was unknown in 2 patients.

The event-free survival rate for the primary endpoint was 93.2% in the total GP group. Univariable Cox regression analysis revealed that age, history of a malignancy or stroke, not using an ACE-inhibitor/angiotensin-II (AT2) antagonist or aspirin, an impaired LVEF, an MR grade ≥2 and multivessel disease during pPCI were significant predictors for 5‑year all-cause mortality. After multivariable analysis, age, not using an ACE-inhibitor/AT2-antagonist and an impaired LVEF remained statistically significant predictors for the primary endpoint (Tab. 2). The GP patients were stratified by high risk and low risk. Fig. 2 shows that high-risk GP patients (n = 417) have a significant lower event-free survival rate of 88.6% compared with 97.4% in the low-risk GP group (n = 416) (log rank <0.001).

In total, 147 reached the secondary endpoint. A recurrent MI occurred in 36 patients, in 51 cases a patient was revascularised, 42 patients died, 15 patients had a cerebrovascular event, in 2 patients an ICD or pacemaker was implanted and 1 patient was admitted for heart failure. In total, 80.2% remained event-free after 5 years for the secondary endpoint. Tab. 3 demonstrates the univariable and multivariable predictors for MACE within 5 years. Patients with an unfavourable outcome according to the univariate Cox regression analysis were patients who (1) were older; (2) had a current smoking status; (3) were not using aspirin; (4) had a lower LVEF; (5) had an MR grade ≥2; and (6) who had a multivessel disease during pPCI. Patients who underwent complete revascularisation during pPCI had a favourable outcome in the univariate analysis. Current smoking status and an MR grade ≥2 remained significant predictors for MACE after multivariable Cox regression analysis. Fig. 3showed the Kaplan-Meier curves of the patients who were stratified by high- and low risk GP patients. High-risk GP patients (n = 387) reached the secondary endpoint in 73.8% of cases, compared with 88.3% in the low-risk GP group (n = 387) (log-rank <0.001).