Impact of acute total occlusion of the culprit artery on outcome in NSTEMI based on the results of a large national registry

Background The impact of acute total occlusion (TO) of the culprit artery in non-ST-segment elevation myocardial infarction (NSTEMI) is not fully established. We aimed to evaluate the clinical and angiographic phenotype and outcome of NSTEMI patients with TO (NSTEMITO) compared to NSTEMI patients without TO (NSTEMINTO) and those with ST-segment elevation and TO (STEMITO). Methods Demographic, clinical and procedure-related data of patients with acute myocardial infarction who underwent percutaneous coronary intervention (PCI) between 2014 and 2017 from the Polish National Registry were analysed. Results We evaluated 131,729 patients: NSTEMINTO (n = 65,206), NSTEMITO (n = 16,209) and STEMITO (n = 50,314). The NSTEMITO group had intermediate results compared to the NSTEMINTO and STEMITO groups regarding mean age (68.78 ± 11.39 vs 65.98 ± 11.61 vs 64.86 ± 12.04 (years), p < 0.0001), Killip class IV on admission (1.69 vs 2.48 vs 5.03 (%), p < 0.0001), cardiac arrest before admission (2.19 vs 3.09 vs 6.02 (%), p < 0.0001) and death during PCI (0.43 vs 0.97 vs 1.76 (%), p < 0.0001)—for NSTEMINTO, NSTEMITO and STEMITO, respectively. However, we noticed that the NSTEMITO group had the longest time from pain to first medical contact (median 4.0 vs 5.0 vs 2.0 (hours), p < 0.0001) and the lowest frequency of TIMI flow grade 3 after PCI (88.61 vs 83.36 vs 95.57 (%), p < 0.0001) and that the left circumflex artery (LCx) was most often the culprit lesion (14.09 vs 35.86 vs 25.42 (%), p < 0.0001). Conclusions The NSTEMITO group clearly differed from the NSTEMINTO group. NSTEMITO appears to be an intermediate condition between NSTEMINTO and STEMITO, although NSTEMITO patients have the longest time delay to and the worst result of PCI, which can be explained by the location of the culprit lesion in the LCx.


Background
According to the European Society of Cardiology (ESC) guidelines, patients with myocardial infarction (MI) and ST-segment elevation (STEMI) are eligible for emergency reperfusion therapy, whereas those with non-ST-segment elevation MI (NSTEMI) require further risk stratification; thus, the qualification for invasive diagnosis and treatment is delayed [1]. The STEMI-NSTEMI paradigm Open Access *Correspondence: rajzer37@interia.pl 1 Jagiellonian University Medical College, 1st Department of Cardiology, Interventional Electrocardiology and Arterial Hypertension, Jakubowskiego St. 2, 30-688 Kraków, Poland Full list of author information is available at the end of the article is based on the observation that ST-segment elevation (STE) on the electrocardiogram (ECG) in the majority of patients with MI is associated with acute total occlusion (TO) of the infarct-related artery (IRA), while subtotal IRA occlusion leads mostly to ST-segment depression and negative T-waves on the ECG. However, when qualification for emergency reperfusion therapy is based on ECG criteria, we lose approximately 25% of patients with acute TO of the IRA who do not present STE [2]. NSTEMI patients form a very heterogeneous group, and ESC guidelines recommend urgent coronary angiography only for those with life-threatening ventricular arrhythmias, resistant angina pectoris and haemodynamic instability. These conditions may be accompanied by total IRA occlusion, but this is not always the case. Thus, percutaneous coronary intervention (PCI) may be deferred in a substantial subset of NSTEMI patients with TO of the IRA, which may result in delayed myocardial salvage and poor cardiovascular outcomes [3].
This study aimed to identify the key points of clinical characteristics, course of treatment and outcome of patients with NSTEMI with TO of the IRA (NSTEMI TO ) by comparison with the two most outlying groups: patients with NSTEMI and a non-occluded coronary artery (NSTEMI NTO ) and patients with STE and an occluded IRA (STEMI TO ).

Methods
We analysed the data of patients with MI assembled within 48 months (2014-2017) into the Polish National Database of Invasive Coronary Procedures (ORPKI), coordinated by Jagiellonian University Medical College and endorsed by the Association of Cardiovascular Interventions of the Polish Cardiac Society [4]. All clinical data were collected by the operator and then uploaded into the database after each procedure. The diagnosis of NSTEMI or STEMI, recognition of the IRA, all clinical decisions during the coronary invasive procedure and definitions of periprocedural complications were left to the uploading ORPKI operators' experience and knowledge according to current ESC guidelines.
Acute TO of the IRA was defined in our study as Thrombolysis In Myocardial Infarction (TIMI) 0 flow during coronary angiography in patients with MI [5].
To achieve the aim of the study, we compared 3 groups of patients: NSTEMI TO , NSTEMI NTO and STEMI TO and excluded from the analysis patients with STEMI and non-occluded coronary artery (STEMI NTO ), those without significant coronary artery stenosis, those not treated with PCI and those diagnosed with chronic total occlusion of the IRA. To minimize confounding factors influencing the electrocardiographic presentation of MI, we decided to perform angiographic analysis only for patients with PCI of single native vessel disease.
Our study was an observational, non-experimental, retrospective analysis and was performed in accordance with the relevant guidelines and regulations. Only anonymized data were included in the research analysis, and according to Regulation 2016/679 of the European Parliament and of the Council (EU) from 27 April 2016 on the protection of individuals with regard to the processing of personal data and on the free movement of such data and with art. 9 section 2, this study did not require any additional ethics board approval. All subjects in our study gave informed consent for personal data processing by the Association of Cardiovascular Interventions of the Polish Cardiac Society before percutaneous coronary intervention.

Statistical analysis
Categorical variables are presented as numbers and percentages. Continuous variables are expressed as the mean ± standard deviation (SD) or the median and interquartile range (IQR). The normality of continuous variables was assessed by the Kolmogorov-Smirnov-Lilliefors test. Equality of variance was assessed using Levene's test. Differences between three groups were compared using classic one-way analysis of variance (ANOVA) or Welch's ANOVA depending on the equality of variance for normally distributed variables. The Kruskal-Wallis test was used for ordinal or non-normally distributed continuous variables. Categorical variables were compared by Pearson's chi-square test. All post hoc analyses were performed using the Benjamini-Hochberg procedure for controlling the false discovery rate (FDR). Two-sided p-values < 0.05 were considered statistically significant. All calculations were performed with JMP ® , Version 14.2.0 (SAS Institute Inc., Cary, NC, USA).

Results
The results of 245,869 coronary angiography procedures performed in patients with MI were entered into the ORPKI registry. After exclusion of patients (a) without significant stenosis of coronary arteries, (b) without occlusion of the IRA in STEMI, (c) without PCI treatment; (d) with chronic total occlusion of the IRA and (e) with multivessel PCI treatment, 131,729 patients who underwent single-vessel PCI constituted the study group. Among them, 65,206 (80.09%) patients with NSTEMI had no TO of the IRA (TIMI > 0), while total occlusion of the IRA (TIMI = 0) was found in 16,209 (19.91%) patients with NSTEMI and 50,314 (48.21%) with STEMI. The study flowchart is shown in Fig. 1.

Clinical characteristics of the study groups
Patients with NSTEMI TO were younger than those with NSTEMI NTO but older than patients with STEMI TO . The percentage of smokers was highest in the STEMI TO group, lower in the NSTEMI TO group and lowest in NSTEMI NTO group. The prevalence of chronic diseases (arterial hypertension, diabetes, chronic kidney disease, chronic obstructive pulmonary disease) was highest in the NSTEMI NTO group, lower in the NSTEMI TO group and lowest in the STEMI TO group. All of the aforementioned differences were significant (p < 0.0001). A similar tendency was observed in the history of previous coronary revascularization (PCI or CABG), MI or stroke, which were most common in patients with NSTEMI NTO , less common in those with NSTEMI TO and least common in those with STEMI TO (p < 0.0001).
Clinical status on admission in the NSTEMI TO group was more severe than that in the NSTEMI NTO group but less serious than that in the STEMI TO group. More advanced Killip classes were observed with the highest frequencies in patients with STEMI TO , lower frequencies in those with NSTEMI TO and the lowest frequencies in those with NSTEMI NTO . Cardiac arrest before admission was more common in patients with STEMI TO than in those with NSTEMI TO and more common in patients with NSTEMI TO than in those with NSTEMI NTO (Table 1).

Time delays in MI treatment within study groups
Direct transport to the catherization laboratory (Cath lab) was most common in the STEMI TO group, less common in the NSTEMI TO group and least common in the NSTEMI NTO group. The time from pain to first medical contact (FMC) was longer in the NSTEMI TO group than in either the STEMI TO or NSTEMI NTO group. Time periods (from pain to balloon inflation and from FMC to inflation) were shortest in the STEMI TO group, intermediate in the NSTEMI TO group and longest in the NSTEMI NTO group. Time from FMC to inflation < 90 min and time from FMC to inflation < 120 min were observed most frequently in patients with STEMI TO , less frequently in those with NSTEMI TO and least frequently in those with NSTEMI NTO (Table 2, Fig. 2).

Results of coronary angiography
In our study, the left anterior descending artery (LAD) was identified as the IRA in 45,008 patients; the left circumflex artery (LCx), in 29,479 patients; and the right  In contrast, occlusion of LAD as the culprit lesion was observed the least often in this group. In patients with STEMI TO , LCx occlusion was infrequent, while occlusion    of the RCA or LAD was prevalent. NSTEMI NTO was related most often to LAD as the culprit lesion, less commonly to RCA and least often to LCx (p for contingency analysis < 0.0001).

Analyses of PCI results
The successful revascularization outcome defined as TIMI flow grade after PCI in the NSTEMI TO group was worse than that in the STEMI TO and NSTEMI NTO groups ( Table 3). TIMI flow grade 3 was reached with the lowest frequency in the IRA after PCI, and TIMI flow grade 0 after PCI was noticed with the highest occurrence rate in the NSTEMI TO group compared with both the STEMI TO and NSTEMI NTO groups. The no-reflow phenomenon, cardiac arrest during PCI and death during the invasive procedure in the NSTEMI TO group occurred less frequently than in the STEMI TO group but more frequently than in the NSTEMI NTO group.

Discussion
To the best of our knowledge, we conducted the largest single-study analysis dedicated to the NSTEMI TO phenomenon (16,209 patients). The previous meta-analyses on this topic included 10,415 patients (7 studies) and 17,212 patients (25 studies) with NSTEMI TO [6,7].
Our study results suggest that NSTEMI TO may be considered an intermediate condition between NSTEMI NTO and STEMI TO . However, the following features make the NSTEMI TO group exceptional: • The longest time delay to obtain proper medical care (patients with NSTEMI TO reached FMC when STEMI TO patients had already undergone PCI), • LCx as the most frequent infarct-related artery, and • The worst final result of PCI.
Numerous studies have shown the differences in the baseline clinical presentation between patients with STEMI and NSTEMI. In the OPERA Registry, correlates of mid-and long-term mortality were similar for NSTEMI and STEMI patients [8]. This leads to the conclusion that we should not consider STEMI and NSTEMI to be two different diseases but rather as an ischaemic continuum due to subtotal or total occlusion of the coronary artery with different ECG manifestations [9,10]. Total occlusion of the IRA can occur in STEMI and NSTEMI patients. Numerous studies have compared acute total occlusion of the IRA with non-total occlusion of the IRA but mostly within the NSTEMI subset of patients [11,12]. Our goal was to compare three manifestations of acute MI: NSTEMI NTO , NSTEMI TO and STEMI TO ; thus, for the first time, we compared three groups instead of two.
Considering the baseline characteristics, patients with NSTEMI TO in our study constituted an intermediate group between NSTEMI NTO and STEMI TO . In comparison to STEMI TO patients, NSTEMI TO patients were older and had a higher prevalence of cardiovascular risk factors and chronic diseases. When comparing NSTEMI TO to NSTEMI NTO , the patients were younger and had a lower prevalence of cardiovascular risk factors and chronic diseases. These findings are in accordance with other studies, where patients with NSTEMI, in comparison to STEMI, were older and more often had chronic diseases [13,14]. According to the baseline characteristics, our NSTEMI TO group was definitely closer to the STEMI TO group than to the NSTEMI NTO group. Patients with STEMI NTO were excluded due to high group heterogeneity. To summarize the results of pre-hospital management, participants with NSTEMI TO were generally less frequently considered candidates for direct transportation to the Cath lab than STEMI TO patients (9.41% vs Table 3 Percutaneous coronary intervention results  25.69%). Additionally, ischaemia time, i.e., time from pain to balloon inflation, as well as time from FMC to balloon inflation, were longer in the NSTEMI TO group than in the STEMI TO group. The duration of ischaemia is a major determinant of infarct size and subsequent mortality [3,15]. In almost all studies included in the meta-analysis of Khan et al., patients with NSTEMI TO had a mean delay to invasive procedure longer than 24 h and, in comparison to patients with NSTEMI NTO , an increased risk of both major adverse cardiovascular events and mortality [6]. The mean time from pain to inflation in our study was approximately 30 h (data not presented) and was similar to that presented by Khan et al. [6]. The time from pain to FMC was the longest in the NSTEMI TO group and was even longer than that in the NSTEMI NTO group. In the NSTEMI TO group, patients postponed the decision to seek medical help, probably because of younger age (than in the NSTEMI NTO group) and a lack of previous experience with stenocardial pain. The longer time delay from pain to FMC in NSTEMI TO than in STEMI TO may be explained by the lower severity of symptoms due to the lower degree of ischaemia in the case of LCx occlusion.
The time delay to achieve the opening of the occluded artery in the NSTEMI TO group in comparison to the STEMI TO group was amplified during in-hospital management, which was noticeable as the pronounced difference (median time from FMC to balloon inflation was almost three times longer in the NSTEMI TO group).
In contrast, patients with NSTEMI TO in comparison to those with NSTEMI NTO were previously considered candidates for invasive management. The potential explanation is the more severe clinical presentation caused by total artery occlusion. The higher frequency of cardiac arrest before admission and more advanced Killip class in the NSTEMI TO group than in the NSTEMI NTO group in our study confirm this hypothesis. Similar results were obtained by Shin et al. in the COREA-AMI Registry [16]. Another commonly used parameter of time delay in MI that influences prognosis is the percentage of patients who undergo PCI within 120 min since the onset of symptoms [3,15,17]. In the study of Terkelsen et al., approximately 50% of STEMI patients had balloon inflation within 120 min [17]. In our study, almost 70% of STEMI TO patients, but only 25% of NSTEMI TO patients, had PCI within 120 min.
Approximately 20% of our NSTEMI patients had acute coronary artery occlusion, which is less than that previously reported by Khan (25.5%) and Hung (34%) [6,7]. This difference may be explained by the fact that we used a stricter definition of NSTEMI TO , analysing only patients with TIMI 0 flow, whereas Khan and Hung included patients with TIMI 0-1. Previous studies examining the distribution of occluded arteries in NSTEMI TO patients indicated that the RCA or LCx was the artery most responsible for NSTEMI TO [6,7]. In our study, we found that the LCx was the most typical localization of the culprit lesion responsible for MI in the NSTEMI TO group. The distribution of the IRA differs between trials when STEMI cases are compared to NSTEMI, i.e., in STEMI, there is underrepresentation of the LCx as the IRA [18], whereas in NSTEMI TO, occlusion of the LAD occurs the least often [7,19]. We must acknowledge that ECG has unsatisfactory sensitivity for diagnosis of coronary artery total occlusion, especially in posterolateral distribution [20].
In our study, patients with NSTEMI TO demonstrated a more severe clinical condition on admission than those with NSTEMI NTO (more advanced Killip class, higher prevalence of death and cardiac arrest prior to admission or during an invasive procedure, no-reflow phenomenon), which is in concordance with prior studies showing that the prognosis of patients with total occlusion without ST-segment elevation is worse than that of NSTEMI NTO patients [6,7]. We confirmed that the outcome after PCI (lower frequency of achieving TIMI 3 and higher frequency of TIMI 0) in NSTEMI TO patients is even inferior to that in STEMI TO patients. A possible explanation is that unrecognized acute coronary artery occlusion is associated with high morbidity and mortality [15], and the outcome in this group is worse than that in the group that received timely revascularization [11,21].
In summary, we must acknowledge that the identification of NSTEMI TO patients prior to coronary angiography remains challenging. Among the NSTEMI patients in our study, younger patients with a lower frequency of comorbidities (which is not a typical NSTEMI group characteristic) should be highly suspected of having acute total occlusion of the infarct-related artery if they present severe symptoms on admission (i.e., advanced Killip class, pre-hospital cardiac arrest, long pain duration), which enables us classify them into the very high-risk group according to the most current 2020 ESC NSTEMI guidelines [22]. To select high-risk patients who require urgent coronary intervention, we should exert additional effort and use all available methods, i.e., careful clinical assessment, ECG analysis including additional ECG leads (i.e., V7-V9) and patterns that are highly suggestive of TO of the IRA (extensively described in the supplementary table of 2020 ESC NSTEMI guidelines), specific risk score calculation and echocardiography examination with wall motion and strain analysis [22].

Study limitations
Our study has several limitations. First, we should be very cautious about drawing conclusions about detailed in-hospital prognoses because our analysis is based on data from the structured registry that included prespecified clinical and a periprocedural data spectrum only, without longitudinal follow-up, but with the largest number of evaluated patients thus far. Second, the registry was created, and data were entered by several operators; the quality of data depends on their individual knowledge. However, only the most experienced operators collected the data.

Conclusions
Approximately one-fifth of NSTEMI patients had acute total coronary artery occlusion (NSTEMI TO ). According to the clinical characteristics, NSTEMI TO seems to be an intermediate condition between NSTEMI NTO and STEMI TO . However, it should be emphasized that NSTEMI TO patients have the longest time delay to PCI and the worst final result of PCI, which can be at least partially explained by the most common location of the culprit lesion in the LCx. Therefore, patients with NSTEMI should undergo strict evaluation for signs indicating possible acute total coronary artery occlusion (e.g., younger age, lower cardiovascular risk, fewer chronic diseases but more severe clinical presentation on admission) and should undergo comprehensive ECG and echocardiographic assessment to prevent a time delay to and improve the results of revascularization.