Primary data for this study were obtained by analyzing consecutive patients admitted to our center with STEMI to undergo PCI between 2013 and 2022. The data were originally collected using a written form, and subsequently time differences were coded into a database. All data were anonymized for the purpose of the present analysis.

Demographic data (age and gender) were collected for each patient. To characterize the timeline of the STEMI pathway, data were collected regarding the year that the procedure was performed, time before patients sought medical attention (patient delay), time from FMC to ECG acquisition (ECG delay), time from ECG acquisition to activation of the intervention team (diagnostic delay), time from FMC to diagnosis, time from intervention team activation to departure from the secondary location (logistic delay), time from FMC to departure from the secondary location (door-in door-out [DIDO] time), transport time, time from arrival at the PCI center to arrival in the catheterization laboratory (center delay), procedure time, time from arrival at the PCI center to wire crossing (door-to-wire time), time from diagnosis to wire crossing, time from FMC to wire crossing, and total ischemia time. Figure 1 illustrates the timeline analyzed. Regarding the place of presentation, patients were divided into four groups: coming from primary care physicians (PCPs), emergency first responders (EFRs), our PCI center, or other, non-PCI, centers. The variables of logistic delay and transport time are not applicable to the PCI center population. The variable DIDO is not applicable to the EFR or PCI center groups.

Figure 1.

Visual representation of the timeline of the ST-elevation myocardial infarction pathway. ECG: electrocardiogram; FMC: first medical contact.

(0.14MB).

Normality was verified by the Kolmogorov–Smirnov test. All continuous variables are described using median±interquartile range. Categorical variables are summarized as absolute and relative frequencies for each data category. The Mann–Whitney U test was used to compare numerical variables for the DIDO variable, and the Kruskal–Wallis test and pairwise comparisons using the Bonferroni correction were performed to investigate the specific differences between groups for the other numerical variables. Categorical variables were compared with the chi-square test, and post-hoc analysis using residual analysis was conducted to explore specific group differences. A significance level of 0.05 was assumed for all analyses. The statistical analysis was performed using IBM SPSS version 28.

In the overall cohort, delay from FMC to diagnosis was 92.0±142.0 min There was a significant difference between the origin groups (chi-square=103.5, degrees of freedom [df]=3, p<0.001). EFRs demonstrated a significant difference compared to the PCI center, PCPs and non-PCI centers (z=99.9, p=0.028; z=234.3, p<0.001; and z=309.9, p<0.001), indicating that the PCI center, PCPs and non-PCI centers had significantly higher FMC to diagnosis time compared to EFRs. The PCI center also demonstrated a significant difference compared to PCPs and non-PCI centers (z=134.5, p=0.001; and z=210.0, p<0.001), indicating that PCPs and non-PCI centers had significantly longer FMC to diagnosis time than the PCI center. No significant differences were found between PCPs and non-PCI centers (z=75.5, p=0.123). Median times are represented in Table 2.

The observed median time from STEMI diagnosis to wire crossing was 89.0±58.0 min. There was a significant difference between the origin groups (chi-square=167.8, df=3, p<0.001). The PCI center demonstrated a significant difference compared to EFRs, PCPs and non-PCI centers (z=177.7, p<0.001; z=219.0, p<0.001; and z=399.9, p<0.001), indicating that EFRs, PCPs and non-PCI centers had significantly shorter STEMI diagnosis to wire-crossing time compared to the PCI center. These differences are clearly represented in Figure 2. Non-PCI centers also demonstrated a significant difference compared to PCPs and EFRs (z=222.2, p<0.001; and z=181.0, p<0.001), indicating that PCPs and EFRs had significantly shorter STEMI diagnosis to wire-crossing time than non-PCI centers. No significant differences were found between PCPs and EFRs (z=41.3, p=0.848). Median times are represented in Table 2.

Figure 2.

Boxplot distribution of time from diagnosis to wire crossing, according to place of origin and year. The red line through the y axis marks the 120-min mark. PCI: percutaneous coronary intervention.

(0.22MB).

Only 37 patients (2.2%) from our cohort met the target time from FMC to diagnosis of 10 min. Group analysis showed that there were no significant differences between groups according to origin (chi-square=6.2, df=3, p=0.103). EFRs and the PCI center had the highest proportion of patients who met this target, with 3.2% and 2.9% respectively (Figure 3).

Figure 3.

Graphical representation of proportions of patients who met the 10-min target from first medical contact to diagnosis. FMC: first medical contact; PCI: percutaneous coronary intervention.

(0.11MB).

Around three quarters of the patients (967, 74.1%) met the target in the guidelines for the maximum expected delay from STEMI diagnosis to wire crossing (120 min). There was a significant difference between the origin groups (chi-square=86.5, df=3, p<0.001). Post-hoc analysis showed a significant difference in the distribution in the PCI center compared to non-PCI centers and PCPs (residual=6.2, p<0.050; residual=1.8, p<0.050), indicating that the PCI center had a significantly higher proportion who met the target compared to non-PCI centers and PCPs (86.0% compared to 55.0% and 76.5%, respectively). A significant difference between non-PCI centers and EFRs/PCPs was also found (residual=−5.1, p<0.05; residual=−4.4, p<0.050), indicating that a significantly higher proportion of patients originating from EFRs and PCPs met the target compared to non-PCI center patients. No significant differences were found between the PCI center and EFRs (residual=1.1, p>0.050) or between EFRs and PCPs (residual=0.7, p>0.050) (Figure 4).

Figure 4.

Graphic representation of proportion of patients who met the 120-min target from diagnosis to wire, according to place of origin. PCI: percutaneous coronary intervention.

(0.14MB).

A maximum of 90 min from STEMI diagnosis to wire crossing was achieved in 44.8% of patients not admitted to the PCI center. There was a significant difference between the origin groups (chi-square=43.5, df=2, p<0.001). Post-hoc analysis showed a significant difference in the distribution of non-PCI centers compared to EFRs and PCPs (residual=−2.1, p<0.050; residual=−2.3, p<0.050), indicating that a significantly lower proportion of patients who met the target came from non-PCI centers compared to EFR and PCP patients (29.4% vs. 52.2% and 51.1%, respectively). No significant difference was found between PCPs and EFRs (residual=−0.2, p>0.050) (Figure 5).

Figure 5.

Graphic representation of proportion of patients who met the 90-min target from diagnosis to wire for transferred patients, according to place of origin. Statistically significant differences are shown in red above the graph. PCI: percutaneous coronary intervention.

(0.11MB).

In the PCI center group, a maximum of 60 min from STEMI diagnosis to wire crossing was achieved in 40.6% of patients.

Median patient delay was 92.0±146.0 min and there was no significant difference between the origin groups (chi-square=3.9, df=3, p=0.274). The procedure duration median was 28.0±13.5 min, and no significant differences were found between origin groups (chi-square=0.391, df=3, p=0.942).

ECG delay was 19.0±146.0 min and there were significant differences between origin groups (chi-square=78.3, df=3, p<0.001). EFRs demonstrated a significant difference compared to the PCI center, PCPs and non-PCI centers (z=181.2, p<0.001; z=212.3, p<0.001; and z=290.0, p<0.001, respectively), indicating that there was significantly greater ECG delay in the PCI center, PCPs and non-PCI centers than in EFRs (21.0 min, 21.0 min and 28.0 min, respectively vs. 10.0 min). There was also a significant difference between the PCI center and non-PCI centers (z=105.8, p=0.020), indicating that non-PCI centers had significantly greater ECG delays than the PCI center. No significant differences were found between PCPs and non-PCI centers (z=77.7, p=0.122) or the PCI center and PCPs (z=28.1, p=1.000).

The median logistic delay associated with patient transfer was 15.5±46.3 min. There were significant differences between origin groups (chi-square=36.2, df=2, p<0.001), including EFRs compared to PCPs and non-PCI centers (z=42.5, p=0.028; z=89.2, p<0.001), indicating that logistic delays were significantly longer for PCPs and non-PCI centers compared to EFRs (11.5 and 35.0 min, respectively, vs. 5.0 min). There was also a significant difference between PCPs and non-PCI centers (z=46.7, p=0.003), indicating that PCPs were associated with a significantly shorter logistic delay than non-PCI centers.

Over the period between 2013 and 2022, there were no significant differences in STEMI diagnosis to wire-crossing time (chi-square=12.9, df=9, p=0.169). Regarding origin groups, no statistical difference over the years was found in the EFR group (chi-square=8.367, df=9, p=0.498), but there were significant differences over the years in all of the other groups (PCPs: chi-square=29.7, df=9, p<0.001; non-PCI centers: chi-square=32.9, df=9, p=<0.001; PCI center: chi-square=33.0, df=9, p<0.001). These results are represented schematically in Figure 2.