Acute Coronary Syndromes and Inflammatory Bowel Disease: The Gut–Heart Connection

(1) Background: Inflammatory bowel disease (IBD) induces a process of systemic inflammation, sharing common ground with acute coronary syndromes (ACS). Growing evidence points towards a possible association between IBD and an increased risk of ACS, yet the topic is still inconclusive. Therefore, we conducted a systematic review aiming to clarify these gaps in the evidence. (2) Methods: We conducted a systematic search on EMBASE, Cochrane Library, and PubMed, identifying observational studies published prior to November 2020. The diagnosis of IBD was confirmed via histopathology or codes. Full articles that fulfilled our criteria were included. Quality assessment was performed using the Newcastle–Ottawa scale (NOS). (3) Results: We included twenty observational studies with a total population of ~132 million subjects. Fifteen studies reported a significant association between ACS and IBD, while the remaining five studies reported no increase in ACS risk in IBD patients. (4) Conclusions: ACS risk in IBD patients is related to hospitalizations, acute active flares, periods of active disease, and complications, with a risk reduction during remission. Interestingly, a general increase in ACS risk was reported in younger IBD patients. The role of corticosteroids and oral contraceptive pills in increasing the ACS risk of IBD patients should be investigated.


Introduction
Inflammatory bowel disease (IBD), comprising Crohn's disease (CD) and Ulcerative colitis (UC), outlines the incurable chronic inflammation of the gastrointestinal tract affecting approximately 2.2 million people in Europe [1] and 7.7 million Americans [2], whilst in Asia the incidence is 1.4 per 100,000 [3]. Nevertheless, the discrepancies in healthcare infrastructure and epidemiological reporting should be considered.
The number of patients with IBD is growing exponentially and is expected to significantly increase in the western world. Previously, IBD was thought to be limited to Caucasians in western countries but not anymore, as IBD was found to be rather related to environmental factors than ethnicity or heredity since most people do not present a family history and twin studies have not proven any concordance [4]. In addition, the incidence of IBD differs between regions in which the genetic background is similar. An increase in IBD incidence and prevalence in newly industrialized countries is being explained by the populations' shift towards urbanism, lifestyle changes such as smoking and diet, and increased exposure to pollution and sedentarism [5].
CD and UC mostly affect young adults and adolescents between 20-30 and 30-40 years, respectively, although they have been reported to show a second peak at 60-70, showing a bi-modal distribution with the incidence of UC twice that of CD [5]. Smoking and appendectomy are risk factors demonstrated to affect the IBD risk [6]. Taking into consideration the demographics most affected by this condition and the relapsing course of this disease, chronic management is necessary.
Although not yet fully understood, IBD pathogenesis involves a series of pathologic immune-mediated processes in individuals with a genetic predisposition [7].
Inflammation, being a key process in IBD, is also involved at all levels of coronary atherosclerosis and acute coronary syndromes (ACS), from the initial plaque formation to the thrombus rupture. It is worth noting the well-established association between IBD and the increased risk of venous thromboembolism due to the pro-atherogenic nature of the disease, especially during active flares [1].
Recently, several studies have cast light on the possible association between IBD and the increased risk of ACS. However, these assertions remain disputed as other studies claim no association. Therefore, we conducted the first systematic review to the best of our knowledge evaluating whether IBD patients are associated with an increased risk of ACS.

Materials and Methods
This systematic review was written following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines [9].

Data Sources and Search Strategy
The electronic databases PubMed, EMBASE, and Cochrane Library were searched without any restrictions from their inception until November 4th 2020 to identify potential observational studies. The following search string was entered for PubMed (("Acute Coro- , and similar search terms were used for EMBASE and Cochrane Library. Furthermore, we manually sorted the pertinent results across the three databases with the purpose of reducing results bias.

Study Selection and Eligibility Criteria
Observational studies assessing the association between ACS and IBD were eligible for inclusion. The original articles were included in the systematic review and qualitive assessment if they satisfied the following criteria: (1) Observational-study, population/hospital/primary care-based; (2) Inflammatory bowel disease confirmed by histopathology or ICD codes; (3) ACS diagnosis based on the criteria established in each study; (4) Studies on humans solely.
Exclusion criteria: (1) Experimental studies; (2) Studies published in languages other than English, German, or Romanian; (3) Case reports, letters, reviews, short surveys, practice guidelines, press articles, conference abstracts/papers; (4) Abstracts published without full-text or with the paper unavailable.
Two investigators (A.J. and A.I.) evaluated the titles and abstracts so that studies satisfying the inclusion and exclusion criteria were further assessed by reviewing the full paper, while in the case of discrepancies between the two investigators, a consensus was reached through discussion. Irrelevant studies were excluded.

Data Extraction
We extracted the following data from the included studies: author's name, publication year, country, total subjects, study population, ACS patients, ACS mortality, IBD patients including CD and UC, mean age, gender distribution, IBD severity, treatment, follow-up duration, and the main findings. Data were extracted and entered by A.J while S.L.P. reviewed the extracted data for possible inaccuracies. Any discrepancies regarding the outcome of the data extraction were resolved through discussion. Extracted data were entered in a Spreadsheet while the final data were aggregated into the presented manuscript.

Quality Assessment
The investigators (A.J. and S.L.P.) independently used the Newcastle-Ottawa scale (NOS) in order to objectively assess the bias risk and internal validity of the included studies [10]. Separate assessment forms were used for case-control studies, cohort studies, and cross-sectional studies. Studies were scored based on how many stars were obtained and criteria were verified in the selection, comparability, and outcome section and thereafter graded accordingly following the evaluation, with scores ranging between 0 and 9 stars. Studies that received 7 stars or more were of high quality. The number of stars were added up in each study in order to compare the quality of included studies in a quantitative manner. Any discrepancies between the two investigators regarding the quality assessment of the included studies were discussed and a consensus was reached. Figure 1 outlines the PRISMA flow diagram describing the identification, screening, and inclusion phases. The search strategy identified 583 articles throughout the three databases, of which 150 results were from PubMed, 393 from EMBASE, and 40 from Cochrane Library. A total of 93 duplicates were identified and removed. After a title and abstract review, we excluded 422 articles after first assessment. A total of 68 articles were retained for full-text evaluation according to the inclusion and exclusion criteria, of which 48 were excluded due to the following reasons:

Study Characteristics
The summarized characteristics of the 20 included studies are shown in Table 1. About 132 million subjects are included in this study where IBD cases ranged from 80 to 563,687.

Study Characteristics
The summarized characteristics of the 20 included studies are shown in Table 1. About 132 million subjects are included in this study where IBD cases ranged from 80 to 563,687.
Five studies were conducted in Europe (France n = 1, Greece n = 1, Denmark n = 2, UK n = 1), two studies in Asia (Taiwan n = 1, South Korea n = 1), and thirteen in North America (USA n = 13).  Comorbidities and complications more frequently associated with IBD were strongly associated with mortality from ACS. Among IBD patients with ACS, comorbidities and complications that were potentially related to IBD were strong independent predictors of increased mortality.  • ACS mortality has a death parameter but associated with CAD not ACS specifically. Patients with IBD are diagnosed with CAD at a younger age as compared with non-IBD patients, are less likely to be active smokers and have lower body mass index. There was no difference in post-PCI major adverse cardiovascular outcomes.

Definition of IBD
Most studies confirmed the diagnosis of IBD using colonoscopy and histopathological examination or the International Classification of Disease (ICD) codes.

IBD Potentially Increasing the Odds of ACS Occurrence
Several studies evaluated whether IBD increases the risk of a patient developing ACS, with inconsistent results.
Mendelsohn et al. assessed 80 Crohn's patients where 12% of the deaths were associated with myocardial infarction (MI) while suggesting that it might have been aggravated by steroid therapy [59]. A retrospective cohort analysis by Ha et al. evaluated 17,487 patients with IBD, concluding that women between 40-59 showed an increased risk of MI in comparison to their age matched groups [61]. Moreover, Kristensen et al. conducted two retrospective cohort studies, the first study involving patients > 15 years old with IBD and dispensed treatment. The authors reported that IBD patients were at a higher risk (two-fold) of MI, stroke, and cardiovascular mortality during episodes of flares and active disease but not during remission phases. The second study by Kristensen et al. involved 73,451 patients ≥ 30 years of age who were hospitalized for the first time for MI. The result was that patients with IBD, especially during flares, were shown to have a high risk of mortality from major adverse cardiovascular events after an MI [1,64]. Furthermore, Aggarwal et al. evaluated 131 patients with IBD, diagnosed with CAD via catheterization, demonstrating that IBD patients were diagnosed with CAD at a younger age while there was no major difference in post-PCI adverse outcomes [65]. A retrospective cohort study from Taiwan by Tsai et al. studied 11,822 patients with IBD symptoms [66] and found that IBD patients were more likely to present the traditional risk factors for ACS and that patients requiring two or more hospitalizations were almost 20-fold more likely to endure an ACS event than those requiring one hospitalization per year, and that IBD women were more likely to have an ACS in the adjusted HR, which further supports the results found by Kristensen et al. [64]. A retrospective cross-sectional study by Kuy et al. studied 461,415 IBD patients and reported an alarming increase in embolic events in these patients [67]. Barnes et al. conducted a nation-wide retrospective cross-sectional study with 78,684,687 subjects, demonstrating that IBD patients were younger and have 0.51-fold odds of having an MI compared to non-IBD subjects, while it was more likely to occur in UC compared to CD [69]. A population-based cohort study by Aniwan et al. involving 736 IBD patients in Olmsted County, Minnesota, reported an ACS percentage of 10.19% and a significant increase in relative risk of acute MI in these patients, with an even amplified risk with systemic corticosteroids [71]. The Le Gall et al. case-control study conducted on 3539 IBD patients suggested that the disease activity is an independent factor in the risk of an acute arterial event [22]. A South Korean cohort study by Choi et al. with 37,477 IBD patients demonstrated a higher risk of MI in CD, with a stronger trend in females [7]. Panhwar et al. conducted a retrospective cohort analysis with 29,090,220 participants, which concluded with a significantly higher risk of MI in younger IBD patients compared to non-IBD patients [2]. A retrospective cohort conducted by Card et al. studied 31,175 patients, reporting an increased hazard of MI in ambulatory active IBD and a lower hazard of MI in hospitalized patients [72].
On the other hand, Pemmasani et al. reported in their retrospective cohort study involving 6,896,635 ACS patients that patients with concurrent IBD had a modestly favorable CVD risk factor profile. However, the mortality from ACS was strongly associated with IBD complications, which are potentially considered strong independent predictors of increased mortality [73]. Moreover, Osterman et al. conducted a retrospective cohort analysis demonstrating that UC and CD patients were not at an increased risk of MI, unlike other chronic inflammatory diseases (RA, SLE, psoriasis, etc.) [62]. The cohort study performed by Merril et al. comprising 271,368 patients with IBD undergoing surgery concluded that there was no association between IBD and perioperative MI and stroke [63]. A case-control study conducted by Ehrenpreis et al. included 5349 patients with code for acute MI, pneumonia, and congestive heart failure with a co-diagnosis of IBD, reported that IBD confers survival benefits for patients hospitalized with acute MI and 34% survival in patients hospitalized for AMI with a CD co-diagnosis [70]. The retrospective cohort study with 3,917,894 patients where IBD percentage was 0.39% and ACS percentage among IBD patients was 2% conducted by Gauravpal et al. [75] demonstrated a statistically non-significant difference in the incidence between IBD and non-IBD patients, with the cardiovascular mortality being 2.3% and 2.1%, respectively.

Discussion
The issue of whether IBD increases the risk of ACS has been addressed in several studies with conflicting results. Our systematic review addresses this issue, as it includes twenty articles with a total study population of approximately 132 million individuals, out of which, fifteen studies reported an increased risk of ACS in IBD patients. All but two of these fifteen studies were rated with a score of >7 stars on quality assessment using NOS, while the remaining two received 5-6 stars on the quality assessment [65,67]. All five studies reporting no significant increase in ACS risk in IBD patients received >7 stars. Furthermore, the studies that reported a significant association between ACS and IBD nine of them were cohort studies, three cross-sectional and one case-control. On the other hand, among the studies that reported no significant association four were cohort studies and one case-control.
In our systematic review, several findings need to be elaborated. Several studies associated an increased risk of MI and morality in ambulatory IBD patients, coupled with other studies that claim a lower percentage of ACS in hospitalized IBD patients. In addition, the study by Tsai et al. reported that patients who needed two or more hospitalizations on average were 20-fold more likely to have ACS than those who required one hospitalization per year. Similarly, Barnes et al. found that IBD patients had 0.51-fold odds of having an acute MI as opposed to non-IBD patients. Furthermore, comorbidities and complications were reported to be independent predictors linked with increased risk and mortality from ACS in acute flares and fulminant stages with IBD. On the other hand, the risk of MI and cardiovascular events did not increase during the remission stages of the disease [1,22,64,72].
Studies that reported an increased percentage of ACS in young IBD patients had a younger study population in contrast to studies that reported no increased risk of ACS in IBD patients. The study conducted by Ha et al. showed that a high percentage of women that used oral contraceptive pills (OCP) had an elevated risk of MI. It is possible that there is a compounding of the risk brought by contraceptives, especially estrogen containing oral contraceptives [74], with the pro-inflammatory state of IBD. This could possibly explain the increased risk of ACS in young women with IBD.
In fact, the increased risk of ACS and mortality in active disease is consistent due to the role of systemic inflammation in increasing atherosclerosis being consistent with the high lipidic profile in IBD patients. Patients with IBD have an increased production of reactive oxygen species (ROS), increased expression of inflammatory cytokines (TNF-α and IL-6) and antibodies that lead to vascular smooth muscle cell proliferation (VSMC), endothelial dysfunction, and the development of CVD [76]. This was illustrated in a recent study conducted by Hernández-Camba et al. that reported a higher frequency of IBD patients being reclassified into a very-high cardiovascular risk via ultrasound assessment of carotid plaques [77]. Interestingly, IBD patients see lower frequencies of acute MI when compared to other chronic inflammatory diseases such as rheumatoid arthritis and systemic lupus erythematosus, which might perhaps highlight a different pathophysiological mechanism where platelet dysfunction is more involved [69].
Two studies cited steroid therapy in IBD as an explanation for the increased relative risk of ACS [68,71]. However, steroids are used in the management of acute flares of IBD, and as stated earlier, several studies have demonstrated an increased risk of ACS in acute flares and the fulminant and active stages of IBD. It is possible that steroid use is not associated with the increased odds of ACS, but the severity of the disease in which they happen to be used more in, which is acute flares and severe IBD. Although the association of corticosteroid usage in IBD patients and ACS is controversial, it is hypothesized that steroids usage increases CVD risk through several pathways including the sympathetic stimulation of the renin-aldosterone-angiotensin axis [76]. All things considered, further studies should evaluate the benefit of immunosuppressive therapy and further investigate the association between corticosteroids and ACS in IBD patients and more aggressive anti-inflammatory therapies should be explored to reduce atherosclerosis, cardiovascular comorbidities, and mortality.
Furthermore  [66,72]. It is worth noting that six papers of the 20 included in this study discussed ACS mortality in IBD [1,60,68,70,73,75], but none assessed IBD severity using the CDAI for Crohn's disease and DAI score for ulcerative colitis.
Our systematic review has several limitations. Although none of the included studies were of poor quality and only three studies were of fair quality [2,65,67], these ratings should be considered cautiously due to possible methodological bias and flaws. Due to the observational design of the studies included in our systematic review, causality between IBD and increased ACS risk cannot be confirmed. Most studies were of retrospective design, without having a follow-up period. Moreover, none of the studies assessed the severity of IBD, which did not allow us to assess the association between IBD staging and the risk of ACS. In our systematic review, we did not include studies published in the grey literature as they might lead to biases in our conclusions due to possible methodological flaws.
Regardless, our systematic review presents several strengths, including a large study population. The topic of our systematic review is of clinical relevance as the prevalence and incidence of IBD is exponential while the rates of newly industrialized nations are growing in parallel and the global burden of IBD on already limited healthcare systems with substantial therapies have significantly changed over the years, whilst more studies are shedding light on a possible increased risk of ACS in IBD patients. Furthermore, we conducted a comprehensive search, including the current studies evaluating our topic and summarizing the currently available data in a nonbiased manner while pointing out the missing data that need to be further assessed in future studies.

Conclusions
In conclusion, the risk of ACS increases significantly with hospitalization and acute active flares, in addition to prolonged periods of active disease. On the other hand, IBD patients in remission present with a lower risk for ACS. The general increased risk of ACS in young IBD patients, possibly due to corticosteroid use, in addition to the effects of estrogen containing OCPs in young IBD female patients, should be further investigated. The interplay between several risk factors including chronic inflammation, thrombosis, corticosteroid use, lipid and endothelial dysfunction, and gut dysbiosis are likely to play a crucial role in the association between IBD and increased ACS risk. A better understanding of these mechanisms may possibly lead to developing novel therapeutic targets in patients with IBD.
Managing IBD patients with ACS risk should be performed through a multidisciplinary team-based approach, while aiming to induce disease remission. Screening and management of cardiovascular risk factors are required, especially in IBD patients with increased risk. Future research is required to better elucidate the pathophysiological mechanisms behind the increased ACS risk in IBD patients. Moreover, further studies assessing the severity of IBD, aside from hospitalizations or corticosteroid prescriptions as surrogate markers for severity, in addition to the effect of biological agents in hospitalized IBD patients, on the risk of ACS remain necessary.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/jcm10204710/s1, Table S1: The Newcastle-Ottawa Scale (NOS) for assessing the quality of cohort studies, Table S2: The Newcastle-Ottawa Scale (NOS) for assessing the quality of cross-sectional studies, Table S3: The Newcastle-Ottawa Scale (NOS) for assessing the quality of case-control studies.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.