Prognostic Outcomes in Acute Coronary Syndrome Patients Without Standard Modiable Risk Factors: A Multi-Ethnic Study Of 8680 Asian Patients

This study examined the outcomes of ACS in patients without (termed in a multiethnic


Introduction
The control of cardiovascular risk factors is paramount in the prevention of adverse cardiovascular outcomes including stroke and acute coronary syndrome (ACS) 1 . Hence, early identi cation and intervention of standard modi able risk factors (SMuRF), such as hypercholesterolemia 2 , hypertension 3 , smoking 4 and diabetes mellitus 5 , are essential in reducing the atherosclerotic cardiovascular disease risk of all individuals 6-9 and prevention of cardiovascular disease 10 . Recent studies have shown a growing proportion of patients without SMuRF (termed SMuRF-less 11 ) who were previously asymptomatic, presenting with ST-segment elevation myocardial infarction (STEMI) 12 . Their prevalence among patients presenting with STEMI has increased over the past decade from 13% to approximately 27% 13,14 , and these patients have a higher in-hospital mortality compared to patients with at least one SMuRF 14 .
To date, there is a paucity of studies examining the outcomes of this pragmatically challenging group of SMuRF-less patients who present with ACS. They are often overlooked in large clinical trials which rarely report the absence of standard modi able risk factors or are less often recruited into trials targeting at atherosclerotic cardiovascular risk intervention. Despite the increasing focus on this group of patients in the West 14-17 , SMuRF-less cohort has not been described in Asia. In addition, most previous studies are only limited to the subgroup of SMuRF-less patients presenting with STEMI, with only few studies on patients with Non-STEMI (NSTEMI) 15,18,19 . This study will be the rst to describe the prevalence and characteristic of SMuRF-less patients, and compare their outcomes to those with SMuRF in a large diverse Asian population presenting with ACS.

Setting and Design
Consecutive patients presenting with ACS to a major tertiary academic percutaneous coronary intervention (PCI)-capable hospital in Singapore between 1st January 2011 and 31st March 2021 were retrospectively studied. The hospital is part of the Western network that provides PCI services, including round-the-clock primary PCI, to the western region in Singapore 20 . This Western network is a hub-andspoke system that consists of our hospital (hub) and two other spoke hospitals. The patients presented with either STEMI or NSTEMI via the Emergency Department at the hub hospital or via inter-hospital transfer from the two spoke hospitals.
Patients included in the study were at least 18 years of age and presented with ACS. Patients with previous ACS, percutaneous coronary intervention (PCI) or coronary artery bypass graft (CABG) were excluded. SMuRFs 6, 21 were de ned as having at least one of the following cardiovascular risk factors: ex-smoker or current smoker, hypertension, diabetes mellitus, or hypercholesterolemia. Hypertension was de ned as previously diagnosed hypertension, prescribed anti-hypertensives, or newly diagnosed hypertension during the index admission. Diabetes mellitus was de ned as previously diagnosed type 1 or 2 diabetes, prescribed glucose lowering medications, or newly diagnosed diabetes during index admission. Hypercholesterolaemia was de ned as previously diagnosed hypercholesterolaemia, prescribed lipid-lowering therapy, or newly diagnosed hypercholesterolaemia during index admission. As both fasting glucose and acute-phase blood pressure readings are in uenced by neurohormonal response to acute myocardial infarction, these were not incorporated in the de nitions.
Patients were allocated to 2 study groups according to their SMuRF status: (1) SMuRF, de ned as having 1 or more SMuRFs, or (2) SMuRF-less, de ned as the absence of SMuRF.

Data Collection
Data on baseline demographic and clinical characteristics, SMuRF status, previous medical history, clinical status at presentation, angiographic and procedural characteristics, echocardiographic characteristics and medications on discharge were retrospectively collected from the electronic clinical records. Information on in-hospital complications were also retrieved.

Study Outcomes
The follow-up period was 30 days and primary outcome was cardiovascular mortality. Secondary outcomes were in-hospital all-cause mortality, unplanned readmission, cardiogenic shock, heart failure, and stroke. Cardiovascular mortality was de ned as any death due to any cardiovascular causes, and allcause mortality was de ned as death due to any or unexplained causes. Cardiogenic shock was de ned by the presence of persistent hypotension de ned as SBP < 90 mm Hg or mean arterial pressure (MAP) 30 mm Hg below the baseline, cardiac index (< 1.8 L/min/m2 without support or < 2.2 L/min/m2 with support) with adequate or elevated lling pressures (left ventricular end diastolic pressure > 18 mm Hg or right ventricular end diastolic pressure > 10 to 15 mm Hg) 22 . Heart failure was de ned clinically based on development of typical signs and symptoms, with structural and functional cardiac abnormalities 23 .
Statistical analysis was conducted on STATA 16.1 (Statacorp) and IBM SPSS Statistics 25 (SPSS Inc., Chicago, IL, USA). A p-value of ≤ 0.05 was considered statistically signi cant. To compare between baseline characteristics of included patients, either chi square analysis or sher exact test was used to compare categorical and binary variables. Unpaired t-test was used in the analysis of continuous variables. Cardiovascular mortality was assessed in the Fine-Gray model with hazard ratio (HR) to account for competing risk. The issue of competing risk has been well described by Abdel-Qadir et al. 24 In the analysis of binary outcomes including 30-day myocardial infarction, stroke, cardiogenic shock, heart failure and readmission, a generalized linear regression with a log link, gaussian distribution, and robust variance estimator were used to compute the risk ratios (RR) 25 . The risk ratio was preferred due to the ease of interpretation compared to an odds ratio 26 . Multivariable adjustment was performed with age, sex, ethnicity, chronic kidney disease, ACS type (STEMI and NSTEMI), cardiac arrest and presence of left main coronary and/or left anterior descending coronary artery disease included in the model. The Kaplan-Meier survival curves for cardiovascular mortality was constructed from the date of admission up to 30 days. The survival curves were further strati ed according to the sex, the three main Asian ethnicities (Chinese, Malay and Indian), and the ACS type (STEMI and NSTEMI). The study was approved by the local institutional review committee in accordance to the revised Declaration of Helsinki (NHG Research -DSRB: 2021/00089-AMD0001). As the study involved the retrospective analysis of clinically acquired data, the institutional review board waived the need for written patient consent.

Study Cohort Characteristics
Of the 8680 patients with ACS enrolled into the study, 7934 (91.4%) patients were in the SMuRF group and 746 (8.6%) in the SMuRF-less group. The follow-up time was 30 days. Between 2011 and 2021, the yearly prevalence of SMuRF-less patients presenting with ACS uctuated little and ranged from 5.9-10.7% (Fig. 1).
During the index hospitalisation, SMuRF-less patients experienced higher rates of PCI (83.0% versus 79.0% respectively, p = 0.018) and were more likely to have culprit vessel involving the left anterior descending artery (56.6% versus 50.9% respectively), or left main coronary artery (3.4% versus 1.9% respectively) compared to SMuRF patients (p = 0.012). Additionally, SMuRF-less patients had higher prevalence of multivessel disease (1.9% versus 0.9% respectively, p = 0.018). Both groups of patients did not differ in the number of vessels and stents involved in PCI, post-PCI Thrombolysis in Myocardial Infarction (TIMI) ow score, PCI success rate and need for CABG.
SMuRF-less patients also had signi cantly higher rates of ventricular arrhythmias, and inotropic and invasive ventilation support when compared to SMuRF patients. However, ischemic mitral regurgitation was less common in the SMuRF-less than SMuRF patients. The incidences of other in-hospital complications including sepsis, atrial brillation, bleeding events, and acute renal failure were similar between the two groups of patients; and so was the length of hospital stay. On discharge, the SMuRF-less group was less likely to be prescribed ACE-inhibitors (ACE-I) or angiotensin-II receptor blockers (ARBs), βblockers, and statins compared to the SMuRF group.
The Kaplan-Meier curve of 30-day cardiovascular mortality are presented based on study cohort, sex, ACS type and ethnicity (Fig. 2). For the overall study cohort, the cumulative event curves diverged early from the day of ACS presentation indicating higher early mortality in the SMuRF-less group compared to the SMuRF group which was sustained over the 30 The multivariable analysis showed that SMuRF-less patients had higher risk of cardiovascular mortality (HR 1.48, 95% CI 1.09-1.86, p = 0.048), cardiogenic shock (RR: 1.31, 95% CI 1.09-1.52, p = 0.015) and stroke (RR: 2.51, 95% CI 1.67-3.34, p = 0.030) compared to the SMuRF patients despite adjusting for important confounders (Fig. 3). The risk of unplanned readmission (RR: 1.10, 95% CI 0.87-1.39, p = 0.413) and heart failure (RR: 0.82, 95% CI 0.56-1.21, p = 0.326) was similar between both patient groups.

Discussion
This study is the rst to examine the prognostic outcomes of an often-overlooked subset of patients without standard modi able risk factors, in a typically understudied Asian population presenting with ACS. It is also the rst of such study to include both patients with NSTEMI and STEMI. The main ndings of the study are: 1) The prevalence of SMuRF-less patients presenting with ACS in an Asian cohort was 8.6%, with its yearly prevalence relatively constant over the past decade; 2) SMuRF-less patients tend to present in a more critical state compared to SMuRF patients, with higher rates of ventricular arrhythmia, and requirement for inotropic and invasive ventilation support; 3) The adjusted risks of cardiovascular mortality, cardiogenic shock and stroke were signi cantly higher in the SMuRF-less patients compared to SMuRF patients; 4) The signi cantly higher cardiovascular mortality in SMuRF-less patients compared to SMuRF patients was apparent early from presentation and was sustained over the 30-day follow-up period. Such trend was observed in men and STEMI patients, but not in women or NSTEMI patients. Similar trend was also seen across all three Asian ethnicities.  13,14 . Notably, the population of SMuRF-less patients in our Asian cohort was much lower with a prevalence of only 8.6%. The stark difference in the proportion of SMuRF-less patients across the globe might be partly explained by the differences in risk factor identi cation 28 , genetic predisposition 29 , lifestyle factors such as smoking and physical activity 30,31 , and individual country's primary prevention programme 28, 32 . Even with the presence of traditional cardiovascular risk factors, their impacts might vary across different ethnic groups, with stroke being more common among hypertensive patients in Asia and chronic heart disease more prevalent in the West 33 . Despite relatively lower than that seen in the West, the prevalence of SMuRF-less patients in our Asian cohort remains sizeable and warrants further attention to address speci c modi able factors that might predispose Asians to various cardiovascular comorbidities.
Even though the SMuRF-less patients in our cohort were generally younger and had fewer baseline comorbidities, their cardiovascular mortality was higher than those with conventional risk factors. This is consistent with the ndings from previous studies 11, [14][15][16]19 based in the West, and could be partly explained by multiple postulated reasons. Several cardiovascular risk factors, such as serum cholesterol or glycated haemoglobin A1c, have a linear relationship with the risk of cardiovascular morbidity and categorising the patients into binary groups using a standard diagnostic threshold can potentially introduce selection bias by missing out on patients with borderline measurements for certain risk factor that have not reached the diagnostic thresholds. As mentioned earlier, individuals with pre-disease state for various cardiovascular risk factors might also have a higher atherosclerotic cardiovascular risk. Moreover, the role of less well established risk factors such as body mass index, triglyceride concentrations, high-density lipoprotein concentration and sedentary lifestyle, which might also be the potential drivers of atherosclerosis but have not been concomitantly evaluated. Additionally, some recognised risk factors such as abdominal obesity, psychosocial factors, sedentary lifestyle, dietary factors and alcohol consumption are not easily quanti ed and hence their potential impact on the outcome of SMuRF-less patients is not well assessed 34 . Furthermore, as patients with known risk factors are more likely to be on treatment, the ACS severity may have been modi ed by evidence-based therapy used in primary intervention 16 leading to better outcome among the SMuRF patients.
The pathogenesis of atherosclerosis, especially its genetic basis, is also not fully understood. A recent study reported as many as 55 genetic loci that are associated with coronary artery disease, with more than 66% of them not linked to the traditional risk factors 35 . Compared to the patients with SMuRF, more SMuRF-less patients in our study were of Indian ethnicity and had family history of premature coronary artery disease highly suggestive of a genetic predilection to develop atherosclerotic cardiovascular disease. It is plausible that these genetic factors might play a major role in the disease process among SMuRF-less patients leading to onset of disease at a younger age and more advanced disease at presentation with consequent worse prognosis. This raises the possibility that other unknown factors might be involved in the pathogenesis and presentation of ACS among the SMuRF-less patients.
Our study found an increased short-term cardiovascular mortality only in the male SMuRF-less patients, which is in contrary to the observation by Figtree et al 11 which showed that SMuRF-less women with STEMI had an excess of short-term mortality over their men counterparts. Such discrepancy in observation could be partly attributed to constitutional differences in study population including ethnic background and the inclusion of full spectrum of ACS patients in our study as opposed to the subgroup of STEMI patients in Figtree et al's study. Other possible factors include varying extent of delay in patient presentation 36 and hormonal-mediated differences associated with atherosclerotic plaque characteristics 37 . However, our study was not granular enough to explain such differences and future studies are warranted.
Moreover, signi cant mortality difference between our SMuRF and SMuRF-less patients was only observed in the STEMI, but not NSTEMI patients. This is in contrast to a prior study that showed increased mortality in SMuRF-less as compared to SMuRF patients with NSTEMI 19 . One possible reason for this discrepancy is the difference in PCI rates between SMuRF and SMuRF-less patients as described by Roe et al, such that the lower use of invasive procedures in SMuRF-less patients might have contributed to their higher mortality 19 . Hence, similar PCI rates between SMuRF-less and SMuRF patients (53.2% versus 48.9% respectively, p = 0.321) might partly explain the similar mortality between the two groups of NSTEMI patients. Another possible reason is the signi cantly lower mortality events in NSTEMI as compared to STEMI patients (4.1% versus 8.3% respectively, p < 0.001) which might lead to less apparent difference seen between SMuRF and SMuRF-less patients among our NSTEMI patients 38 .
Similar to current literature 11,15,16 , we found that SMuRF-less patients were less likely to be treated with guideline-directed medication including beta-blockers, statins and/or ACE-I or ARBs when compared to the SMuRF patients. Figtree et al 11 has shown that suboptimal prescription rate of ACE-I or ARBs and beta-blockers was directly correlated to a higher mortality among the SMuRF-less patients which is in line with other studies demonstrating the prognostic bene t of early initiation of beta-blocker and ACE-I in patients with ACS [39][40][41] . The reason SMuRF-less patients were less likely to be prescribed with prognostically important medication was unclear but could be related to the false perception that they were of lower cardiac risk. The worse clinical status at presentation, lack of pre-existing hypertension and higher incidence of stroke among the SMuRF-less patients might lead to a poorer hemodynamics which precluded the use of beta-blockers or ACE-I or ARBs. Increased awareness of the paradoxical unfavorable outcome in SMuRF-less patients presenting with ACS should be widely promoted and early initiation of guideline-directed medical therapy among ACS patients remains crucial regardless of the cardiovascular risk factor status.

Clinical Implications
Our ndings raise concerns regarding the unfavourable outcome in SMuRF-less patients presenting with ACS among the Asian population. Such patients are not uncommon and may present in an even worse clinical state than those with one or more standard cardiovascular risk factors. These ameliorate the general sense of complacency that signi cant coronary artery disease is an unlikely health concern in individuals without cardiovascular risk factors. Clinicians need to be aware of this unexplained paradoxical phenomenon, and effective lifestyle and pharmacological intervention need to be optimised in all patients regardless of their SMuRF status. Although lower than that reported in the West, the proportion of SMuRF-less patients in our Asian population remains sizeable indicating that this is a global phenomenon that warrant its due attention by all healthcare systems. More efforts are needed to understand the underlying pathophysiology of atherosclerotic cardiovascular risk factors in SMuRF-less patients, from the onset of atherosclerosis through to its progression and the occurrence of ACS, in order to identify such individuals so that appropriate and timely preventative intervention can be given. Currently, most published studies were limited to short-term outcomes and hence further studies are also needed in order to understand the long-term outcome of SMuRF-less patients with ACS.

Strengths and Limitations
This study is the rst to examine the prognosis of SMuRF-less patients presenting with ACS in a large Asian cohort over a 10-year period. However, this study has its limitations. Firstly, this is a single-centre retrospective observational study which might be affected by unknown confounders and bias. Therefore, causality cannot be deduced from our results. However, such potential bias was mitigated by adjusting for important covariates in the multivariable models and using mortality as the primary study outcome. Secondly, the current method of categorizing patients into SMuRF and SMuRF-less groups might not be ideal, but it is the universal method used by all published studies, and based on local or international diagnostic threshold for each of the SMuRFs. Such thresholds are generally derived based on clinical evidence or expert consensus, and usually form the thresholds for guideline-directed treatment. Thirdly, some recognized atherosclerotic cardiovascular risk factors other than those universally considered as SMuRF are also beyond the scope for evaluation in this study. The retrospective nature of the study did not allow further evaluation of SMuRF-less patients for non-atherosclerotic cause of ACS such as protein C and S de ciency.

Conclusion
Patients presenting with ACS but without any standard modi able risk factor are not uncommon in a multiethnic Asian population. They tend to present in a worse clinical state and have poorer short-term outcome, including higher cardiovascular mortality, compared to those with SMuRF.

Declarations
Compliance with Ethical Standards:  Figure 1 Bar graph displaying the prevalence of the SMuRF-less population from 2011 to 2020.  Forest plot comparing unadjusted and adjusted study outcomes in SMuRF-less and SMuRF patients presenting with acute coronary syndrome.