Chronic kidney disease (CKD) is well-known to be a global public health problem and is considered as an important independent risk factor for cardiovascular disease (CVD) development (1, 2). Meanwhile, CVD is a principal cause of death in CKD patients (3). Previous research found that up to 30%–40% of patients presenting with an acute coronary syndrome (ACS) could combine with CKD (4). Despite rapid progress in the treatment, ACS patients with CKD are still facing a high risk of unfavorable clinical outcomes. Traditional cardiovascular risk factors, such as diabetes, hypertension, metabolic abnormalities and aging are prevalent in ACS and CKD (5). These factors collectively contribute to endothelial cell damage, which in turn leads to coronary microvascular dysfunction (CMD). Several studies have highlighted that CMD is closely related to the adverse cardiovascular events of ACS and CKD (6–9). Despite its significance, limited research has been conducted on assessing abnormal coronary microcirculatory function or evaluating the predictive value of CMD in this patient population.

There are various approaches to assessing microvascular function, encompassing both non-invasive and invasive methods (10). Non-invasive methods included coronary computed tomography angiography (CCTA), cardiac magnetic resonance (CMR), and positron emission tomography (PET). Among these methods, cardiac PET is currently regarded as the gold standard for non-invasive assessment of coronary microvascular function (11); CMR has advantages of high-resolution and localization, which provides high diagnostic accuracy (12, 13). However, these methods are often constrained by high costs, the impracticality of repeated measures, or concerns about radiation exposure. Invasive methods include coronary angiography, Doppler flow map, and the index of microcirculatory resistance (IMR) (10). Among them, the thermodilution-based IMR is viewed as the gold standard for the invasive methods for its specificity to the microvasculature, greater quantitative precision, and its advantage of being unaffected by hemodynamic epicardial structural changes (14, 15). Many studies have demonstrated that IMR has diagnostic and prognostic values in a variety of clinical diseases (16–18). For instance, the study conducted by Fearon et al. demonstrated an IMR greater than 40 measured in patients with ST-Segment elevation myocardial infarction (STEMI) predicted adverse events, including rehospitalization, heart failure, or death (19). However, the need for a pressure-temperature sensor guide wire, the use of adenosine to achieve maximal hyperemia and the higher costs of IMR restricted its clinical application (20).

The angiography-derived microvascular resistance (AMR), a recently developed parameter, offers a simpler and more rapid assessment without the need for pressure wires and adenosine (21). There have several investigations examining the role of AMR and suggesting AMR as a valuable and reliable tool for diagnostic purposes. At the same time, AMR also has high predictive potential in a variety of clinical situations, including CAD, STEMI and myocardial infarction (MI) with non-obstructive coronary arteries (MINOCA) (14, 22–24). However, studies on the prognostic value of AMR in ACS patients with CKD have not yet been conducted. In addition, few studies assessed AMR in combination with other predictors, such as the GRACE score, which has a high predictive value in ACS patients. This study aimed to explore the connection between AMR and clinical outcomes and the prognosis value of AMRa and the incremental prognostic value of adding AMR into the GRACE score.

This study first investigated the predictive value of AMR in ACS patients with CKD undergoing PCI. Our results suggested that AMR was independently related to MACEs and all-cause mortality in patients with ACS and CKD. The risk of MACEs and all-cause mortality significantly increased when AMR was ≥ 250 mmHg*s/m. Furthermore, the addition of AMR could improve the predictive value of the GRACE score to predict MACEs and all-cause mortality.

CMD is prevalent among ACS and CKD patients and it is proven to be a significant predictor for short- and long-term clinical outcomes (32–34). De Vita et al. found that ACS exhibited considerable coronary dysfunction, involving both an increased coronary microcirculation vasoconstrictor function and a decreased dilator function (35). CMD is also prevalent in CKD patients (36). Chronic kidney disease (CKD) creates a persistent systemic proinflammatory state that drives endothelial dysfunction, ultimately resulting in coronary microcirculatory impairment. This proinflammatory environment in CKD also induces vascular and myocardial adaptations and remodeling, which contribute to the development of vascular aging and calcification, and myocardial fibrosis (37). These consequences can appear as structural anomalies of the coronary microvasculature, which might result in CMD. Moreover, microvascular rarefaction in CKD patients, combined with endothelial dysfunction and reduced myocardial perfusion, in addition to the left ventricular hypertrophy and diastolic dysfunction, lead to compensatory vasodilation of arterioles and resting coronary flow elevation, and diminished coronary circulatory reserve eventually (5). A further study discovered CMD may mediate the impact of CKD on abnormal cardiac function and cardiovascular events in those without evident coronary artery disease (7). However, patients with CKD were excluded or accounted for only a relatively small part of studies. There is minimal evidence to explore the correlation between CMD and poor prognosis in ACS patients with CKD (38).

IMR, a traditional CMD measurement, is based on thermodilution-pressure wire. However, its clinical application is largely limited by the use of hyperemic agents, the requirement for a pressure wire, longer operation time and the higher cost (39). To overcome these barriers, a coronary angiography images-based calculation of microvascular resistance (AMR) has been proposed (39, 40). By the computational fluid dynamics (CPFD) method, AMR could be evaluated within 1 min, and the entire measurement process takes less than 5 min, allowing for the simultaneous identification of microcirculatory dysfunction during angiographic procedures (41, 42). Several research has proven a diagnostic value of AMR in 56 patients with no obstructive coronary arteries, with 84.2% accuracy, 86.1% sensitivity and 81.0% specificity, respectively, and confirmed that AMR could serve as an independent predictor of adverse cardiovascular events (20, 21, 43). Multiple studies have already proven that AMR was strongly related to adverse events and had an outstanding predictive capacity for adverse outcomes in different cardiovascular diseases (14, 23, 24, 44). However, the impact on prognostic of AMR has not been studied for patients with ACS and CKD.

To our knowledge, this is currently the first study to evaluate the relationship between AMR and clinical prognosis among ACS with CKD patients. We discovered patients with CMD had a higher rate of adverse events and exhibited a significantly increased rate of unfavorable prognostic outcomes after accounting for conventional risk factors. It is very noteworthy that the mortality rate increased significantly in ACS with CKD patients with AMR ≥ 250 mmHg*s/m in our study. Considering the high mortality rate in cases of ACS and CKD, our findings remind clinicians to pay more attention to the clinical management of these patients with CMD or AMR ≥ 250 mmHg*s/m. In addition, our findings revealed that adding the AMR to the GRACE score enhanced its predictive ability and accuracy for MACEs and all-cause mortality, as improvements in C-statistics, NRI and IDI. Our study focused on those with ACS and CKD, who are at a greater risk of overt cardiac dysfunction and a poor prognosis. The results of this study provide new views on the predictive value of AMR for patient outcomes. Taken together, AMR may give extra information about high-risk patients, assisting in the management or prevention of adverse outcomes. Meanwhile, AMR has the potential to increase the use of coronary microvascular function assessments while reducing the use of a pressure-temperature wire and technical errors.

In addition, prognostic stratification using AMR could be useful in many cardiovascular diseases, especially cardiovascular diseases with CMD. For example, MINOCA, a special type of MI, features clinical documentation of an acute MI without angiographically evident obstructive coronary artery obstruction (stenosis < 50%) (45). As an important mechanism, CMD plays an important role in patients with symptoms and/or signs of MI and non-obstructive coronary artery disease, including MINOCA (46). A retrospective multicenter cohort study concentrated on patients with MINOCA conducted by Ciliberti et al. revealed that nearly one out of two patients showed atherosclerosis progression, often requiring revascularization (47). Likewise, a large systemic review that included a total of 55,369 suspected MINOCA participants conducted by Simeone et al. has found that MINOCA was fraught with high rates of mortality, high readmission rates, and socioeconomic burden (48). Over the past few years, the incidence of MACEs in MINOCA patients has increased (49). Our study explores the clinical significance of AMR for assessing coronary microvascular function in a specific group of people. Given the importance of coronary microvascular function, AMR has important prognostic implications and potential therapeutic implications. Assessing the AMR in risk stratification in MINOCA patients is of guiding significance in clinical practice, considering the incidence of total adverse events increasing year by year, as well as the potential therapeutic and prognostic implication of the AMR. More research is needed to extend our findings to other patient populations.

This study still has certain limitations. First, this was a single-center retrospective cohort study, which might cause possible recall bias and be affected by lost follow-up. As a single-center study with a relatively small sample size, additional prospective, large-scale, multi-center investigations are needed to verify our findings. Second, not all coronary angiography images were appropriate for analysis, which may cause possible selection bias. Third, we did not quantify myocardial infarction area or other factors affecting the coronary microcirculation such myocardial bridging (50), which limit any further exploration of the association between AMR and prognosis. Fourth, the majority of the follow-ups were done over the phone or with medical records of readmission, which may be impacted by family economic status or the COVID-19 epidemic. Consequently, we lack data on coronary re-examinations one-year post-MI. Finally, this study was limited to the Chinese ACS with CKD patients, which may restrict the applicability of these findings to other races.

Our findings indicated that AMR was independently associated with an increased risk of MACEs and all-cause mortality in patients with ACS and CKD. Furthermore, adding AMR to the GRACE score could improve the predictive value of MACEs and all-cause mortality. More studies are needed to confirm our findings and assess its predictive value in other cardiovascular diseases to increase its clinical usefulness.