Impact of the risk-treatment paradox on in-hospital and long-term outcomes in patients with very high-risk non-ST-elevation acute coronary syndrome

Methods

This observational study was conducted by the department of Cardiac Emergency, Cardiology Research Institute, Tomsk National Research Medical Center, Tomsk, Russian Federation. Patients were retrospectively recruited for this study during 2023–2024. The study was conducted in compliance with the principles of the Declaration of Helsinki and was approved by the Human Research Ethics Committee of the Institute. The study included 340 patients (aged >18 yr) of both sexes, hospitalized in the emergency cardiology department from January 2019 to June 2021 with a diagnosis of NSTE-ACS.

The inclusion criterion was the presence of at least one very high-risk feature of NSTE-ACS. The very high-risk criteria were assessed retrospectively in each patient according to the updated European Society of Cardiology guidelines (2023)⁷, based on the data from the case records: (i) Hemodynamic instability or cardiogenic shock; (ii) ongoing or recurrent chest pain refractory to pharmacotherapy; (iii) life-threatening arrhythmias or cardiac arrest; (iv) mechanical complications of myocardial infarction; (v) acute heart failure; and (vi) recurrent dynamic ST-segment deviations or T-wave changes.

The assessment was performed by an independent expert. Very high-risk criteria were evaluated at the time of hospital admission, regardless of the invasive strategy chosen during hospitalisation.

RTP was defined as any patient management tactic other than emergency invasive strategy. In the present study, RTP was investigated only from the perspective of invasive strategies. Details of medical treatment were not assessed. Patients received standard treatment for NSTE-ACS according to current clinical guidelines⁷, unless contraindicated.

Ongoing myocardial ischaemia (OMI) was defined as a combination of ongoing/recurrent chest pain and/or dyspnoea with at least one additional established very high-risk criterion. In cases where the patient presented with dyspnoea as the sole clinical symptom against the background of acute heart failure and/or cardiogenic shock (in the absence of chest pain), dynamic ischaemic changes in the ST-T complex on an electrocardiogram were additionally considered to confirm signs of myocardial ischaemia.

Data were stored in a personalized electronic database using Excel 2010 (Certificate of State Registration No. 2023622190 dated 03.07.2023).¹²

Results

The patient cohort (n=340) predominantly consisted of males (185; 54.4%), with a median age of 70 years (62;79). Primary myocardial infarction was diagnosed in 113 patients (33.2%), recurrent myocardial infarction (28 days or more after the previous myocardial infarction) in 116 patients (34.1%), and unstable angina in 111 patients (32.7%).

Invasive coronary angiography (ICA) was performed during the index hospitalization for 277 patients (81.5%); less than half (117; 42%) underwent PCI. The primary reasons for completing ICA as a diagnostic procedure included non-obstructive coronary atherosclerosis (74; 46%), multivessel disease (49; 31%), absence of coronary atherosclerosis (31; 19%), and technical difficulties with PCI (6; 4%), which were related to coronary artery tortuosity and/or severe coronary artery calcification. Subsequently, 21 patients with identified multivessel disease underwent coronary artery bypass grafting (CABG) (42.8%), all later than 24 h after admission, due to the need for preliminary stabilisation and preparation for surgery. CABG was not recommended for 10 patients (20.4%) due to severe comorbid conditions and a high risk of perioperative complications.

ICA within the recommended timeframe (< 2 h) was performed in 127 patients (37.4%). For the remaining cases, ICA was performed later than 2 h after admission (120 cases (35.3%) of early ICA within 2-24 h and 30 cases (8.8%) of delayed ICA within 24-72 h) or not performed (63 cases of conservative treatment; 18.5%), corresponding to RTP situations. Thus, a total of 213 RTP cases (62.6%) were identified.

The main established causes of RTP were chronic kidney disease (stages C3b–C5) (57; 26.7%); medically relieved anginal status in the presence of other signs of very high-risk (40; 18.8%); pulmonary oedema requiring preliminary stabilisation of the condition (28; 13.1%); moderate to severe anaemia (20; 9.4%); and previously known coronary anatomy without data on the presence of an obstructive lesion (18; 8.4%). In 12 patients (5.6%), the cause of RTP was various other clinical conditions, including delirium, nutritional status, infectious conditions, gastrointestinal bleeding, as well as the patient’s refusal to undergo ICA. Furthermore, 32 patients (15%) had combined factors of RTP, while in 6 patients (3%), the cause of RTP could not be determined. The main reasons for the complete refusal of invasive strategy (63 cases; 18.5%) were severe chronic kidney disease and moderate to severe anaemia.

Comparative characteristics of patients according to the presence of RTP are shown in the Table I. Disease outcomes were compared based on RTP presence among all patients (n=340). No significant impact on in-hospital mortality was found (14% vs. 10% in the non-RTP vs. RTP groups, P=0.34; OR 0.71; 95% CI: 0.36–1.37). Evaluation of MACE showed similar achievement rates for secondary endpoints in both groups at 1 month (3.9% vs. 5.2%, P=0.78; OR 1.1; 95% CI: 0.4–3.5) and 1 year (21.2% vs. 18.8%, P= 0.56; OR 0.9; 95% CI: 0.5–1.6) (Table II).

Since ICA did not result in immediate myocardial revascularization in all patients, the outcomes within the PCI subgroups were further analyzed based on the timing of the procedure (within 2 h and >2 h). No statistically significant differences were observed in the frequency of in-hospital mortality or secondary endpoints during long-term follow up (Table II).

Given the heterogeneity of the studied cohort, multivariable regression analysis was performed to identify independent predictors of adverse outcomes. Ongoing myocardial ischaemia at admission emerged as an independent predictor of in-hospital mortality (OR 2.3; 95% CI: 1.9–2.7, P=0.001) and 1-year secondary endpoints occurrence (OR 4.2; 95% CI: 1.14–15.8, P=0.03). RTP did not demonstrate a significant effect on outcomes in the multivariate model. An additional factor associated with the development of 1-year secondary endpoints was ICA performed more than 24 h after admission (24–72 h) (OR 7.5; 95% CI: 1.45–38.3, P =0.016). Early ICA (2–24 h) was a protective factor against the development of 1-year MACE (OR 0.4; 95% CI: 0.16–0.99, P=0.048) in the overall group of very high-risk patients.

In the subgroup of patients with signs of OMI (168; 49.4%), RTP was detected in 94 patients (56%). Significant differences in 3-year secondary endpoints rates were observed, with a higher incidence in the RTP group (29% vs. 47.5%, P=0.02; OR 2.2; 95% CI: 1.1–4.5). Of the 142 patients who underwent ICA in this group, 72 (42.8%) underwent PCI, of which 44 (61.1%) within 2 h and the remaining 28 (38.9%) between 2 and 24 h, corresponding to RTP (Fig. 1).

Close

Fig. 1.

Flowchart for assigning patients to analysing groups. ICA, invasive coronary angiography; PCI, percutaneous coronary intervention.

Export to PPT

Performing emergency PCI compared to early PCI (2–24 h) in the group of patients with OMI reduced the incidence of in-hospital mortality (13.9% vs. 17.8%, P=0.05; OR 1.1; 95% CI: 0.32–4.1) as well as the number of cases achieving the secondary endpoints by the third year of follow-up (10.8% vs. 39.1%, P= 0.009; OR 5.3; 95% CI: 1.4–20.1). Kaplan-Meier analysis further confirmed the differences in the incidence of MACE between these two groups (Fig. 2).

Close

Fig. 2.

Freedom from MACE during three-year follow up in the group of patients with evidence of ongoing myocardial ischaemia according to time of percutaneous coronary intervention (Kaplan-Meier curves, log-rank statistic). MACE, major adverse cardiovascular events.

Export to PPT

In patients without signs of OMI, there was no difference in the incidence of development of secondary endpoints in 3 years between the emergency and early ICA groups (10.3% vs. 14.2%, respectively, P=0.11). In the emergency PCI group, more cases of in-hospital mortality were observed compared to the early PCI group (26.6% vs. 8.7%), but the difference was not statistically significant (P=0.09; OR 0.3; 95% CI: 0.05–2.0).

Discussion

We found that failure to perform emergency ICA within less than 2 h (i.e., the risk-treatment paradox) does not have a negative impact on disease outcomes in the overall group of patients with very high-risk NSTE-ACS in real clinical practice. In the multivariate logistic regression model, early ICA (2–24 h), but not emergency ICA (<2 h), was a protective factor against the development of one-year MACE. Patients with signs of ongoing myocardial ischaemia, determined based on the proposed combination of criteria, received the greatest benefit from emergency ICA aimed at immediate myocardial revascularization, making the implementation of an emergency invasive strategy most justified in this subgroup (Supplementary Figure).

Previous efforts have primarily aimed to determine the most optimal timing for invasive strategies in the general NSTE-ACS population.¹³ Research specifically focused on identifying the benefits of emergency intervention in the very high-risk group remains limited. Our results are consistent with the findings reported by Lupu et al.⁶ The authors observed insufficient adherence to clinical guidelines: only 6.4% of 156 very high-risk patients underwent emergency intervention as recommended by current guidelines. However, despite the limited power of this study, the low frequency of emergency PCI did not result in worse outcomes.

In this study, we defined RTP as the refusal to perform emergency ICA. This approach aligns with previous studies investigating the effectiveness of invasive strategies in patients with NSTE-ACS, with a specific focus on the timing of ICA, generally regardless of the characteristics of coronary lesions.¹³ Our additional analysis showed no significant difference in outcomes between patients with PCI within 2 h and those with PCI after 2 h in the overall very high-risk cohort, similar to the RTP findings.

Two main reasons can be identified to explain why RTP did not have a negative impact on disease outcomes in the study group. First, very high-risk criteria in isolation are highly nonspecific and often mimic signs of acute coronary insufficiency. This is reflected by the high prevalence of non-obstructive coronary artery atherosclerosis in this cohort: 35.4% of patients who underwent emergency ICA had no indications for myocardial revascularization. Second, the expected positive effects of emergency PCI are not fully realized, likely because of the significant influence of comorbid conditions on the patients’ current clinical state. These factors diminish the potential benefits of the recommended emergency ICA.

Data from the Czech CZECH-3 registry¹⁴ have demonstrated that OMI may be an additional predictor of emergency ICA. However, this study included both NSTE-ACS and ST-elevation myocardial infarction patients, which may have confounded the real benefit of immediate intervention. According to our study, patients with signs of OMI at admission who underwent emergency PCI experienced better hospital and three-year outcomes compared to the early invasive treatment group (2–24 h). The obtained data may be explained by greater specificity of signs united under the concept of OMI in verifying myocardial ischaemia compared to using isolated very high-risk criteria. This reflects the need for emergency ICA, aimed at performing PCI in patients with very high-risk NSTE-ACS and OMI signs at hospital admission.

In the group of patients who underwent PCI within the proper timeframe (<2 h) but lacked OMI signs, the rate of in-hospital deaths was unexpectedly higher. These results may have been influenced by concerns regarding the risks associated with ICA and PCI.^15,16 Given these results, it is suggested that, in this very high-risk patient subgroup, a later intervention (up to 24 h) might be feasible, allowing additional time for differential diagnosis and stabilization. However, this conclusion requires further confirmation through larger studies.

This study has several limitations. First, the study is observational, and the retrospective data analysis does not exclude subjectivity during the assessment of very high-risk criteria and unintentional distortion of the patient’s real clinical condition. Second, the study is single centre in nature, resulting in the potential influence of established diagnostic and treatment practices at this institution on patient management decisions.

To conclude, our study did not find a negative impact of the risk-treatment paradox on disease outcomes in the overall group of patients with very high-risk NSTE-ACS. The group of patients with signs of ongoing myocardial ischaemia requires increased attention from clinicians, as such patients benefit most from undergoing ICA with the intention of performing PCI within 2 h.