Risk analysis of gastrointestinal bleeding in hospital patients with acute myocardial infarction undergoing primary PCI

Yan-Yan JIN, Ming YE, Hai GAO

Center for Coronary Artery Disease, Division of Cardiology, Beijing Anzhen Hospital, Capital Medical University, Beijing,China

Dual antiplatelet therapy with aspirin and a P2Y12 inhibitor is the cornerstone of acute myocardial infarction (AMI) management, both invasive and conservative.[1,2]This dual strategy improved ischemic outcomes but was offset by an increased bleeding risk.The prognostic importance of bleeding events has been well established over the past decades, as several studies have shown a strong association between bleeding and mortality.[3]The CRUSADE score is superior to other scores in predicting in-hospital major bleeding events.In this regard, in its non-ST elevation acute coronary syndromes (NSTE-ACS) guidelines, the European Society of Cardiology (ESC) stated that the CRUSADE score could be considered for bleeding risk quantification of coronary angiography in NSTE-ACS patients (class IIb, level B evidence).[4]However, the most common site of spontaneous, non-surgical access site bleeding is the gastrointestinal tract, and gastrointestinal bleeding (GIB) can be prevented to some extent by the prophylactic use of proton pump inhibitors (PPIs),[5]the avoidance of aspirin treatment,[6]or the eradication ofHelicobacter pylori(HP).[7]Nevertheless, research conducted to assess the risk of GIB in patients with AMI is lacking.

The objective of this study was to investigate the incidence of GIB, in-hospital adverse events, and risk factors in acute ST-segment elevation myocardial infarction (STEMI) patients treated with primary percutaneous coronary intervention (PPCI).The data of a total number of 2791 patients with STEMI who underwent PPCI in our hospital from January 2018 to January 2022 were retrospectively analyzed.Of them,1105 patients were excluded according to the exclusion criteria, and 1686 patients were finally included.The patients were divided into a GIB group (248 cases) and a non-GIB group (1438 cases) based on whether GIB was complicated or not during hospitalization.The diagnosis of GIB was defined as brown liquid vomit (positive gastric juice occult blood), black stool, hematochezia, or positive fecal occult blood (≥ 2 times).

The hospital incidence of GIB after STEMI was 11.63%.The in-hospital mortality rate was 1.1% in the STEMI patients receiving PPCI, but was significantly higher in the patients with GIB than in the patients without GIB (3.2%vs.0.7%,P＜ 0.001; Table 1).The level of triiodothyronine (T3) was significantly lower than that in the non-GIb group (1.15 ± 0.31vs.1.20 ±0.33 nmol/L,P= 0.047) and the level of hypersensitive C-reactive protein (hs-CRP) in the GIB group (10.85 ±8.90vs.9.20 ± 8.43 mg/L,P= 0.012) was significantly higher than that in non-GIB group (Table 2).The risk in GIB patients with smoking (59.3%vs.47.5%,P= 0.001) and alcohol consumption (33.5%vs.22.3%,P＜ 0.001) was significantly higher than in those in the non-GIB group (Table 2).Risk factors that may lead to GIB were subjected logistic regression analysis.The results showed that hs-CRP (OR =1.020, 95% CI: 1.001-1.038,P= 0.036), T3 (OR =0.513, 95% CI: 0.297-0.889,P= 0.017), and smoking(OR = 1.494, 95% CI: 1.054-2.118,P= 0.024) were independent risk factors for GIB complications in STEMI patients (Table 3).

Table 1 Comparison of in-hospital outcomes between the two groups.

Table 2 Comparison of clinical conditions between the two groups.

Table 3 Logistics analysis of risk factors for STEMI complicated with gastrointestinal bleeding.

In previous foreign studies, the incidence of GIB in hospitalized patients with AMI was 2.2%-3.9%, and it increased the short-term mortality rate of patients with AMI; the in-hospital mortality rate was 28.2%.[8-9]The incidence of GIB after STEMI decreased from 2.7% in 2003 to 2.0% in 2016 (P＜ 0.001), but the inhospital mortality of patients with GIB was consistently significantly higher than that of non-GIB patients (28.2%vs.11.1%,P＜ 0.001) and persisted over time (P= 0.34).Patients with a GIB complication had a longer hospital stay and their treatment had a higher cost (P＜ 0.001).[9]In this study, we analyzed the data of patients with STEMI who underwent PPCI, and we found that the incidence of inhospital GIB in such patients was 11.63%, which was higher than that in previous studies.[8,9]This difference may be related to the definition of the exclusion criteria applied for patients who did not undergo stool testing during hospitalization.The in-hospital mortality of STEMI patients who received PPCI in the present study was 1.1%, which was significantly lower than that in previous studies,[8,9]but the inhospital mortality of patients in the combined GIB group was still much higher than that in the non-GIB group (3.2%vs.0.7%,P＜ 0.001), which was 4.5 times higher than that in the non-GIB group.The mortality rate of the patients with GIB in this study was significantly lower than that established in previous studies,[8,9]which may be due to the fact that we treated only patients with AMI who received interventional therapy, whereas the patients with AMI enrolled in previous studies included patients with non-invasive treatment such as thrombolysis and drug conservative treatment.These discrepancies may also be related to our exclusion of patients who did not undergo stool testing during hospitalization;probably, patients with severe bleeding did not have the opportunity to undergo stool testing They might also have been associated with the intake of protonpump inhibitors in 99.5% of the patients in our study.Previous investigations have revealed that PPIs therapy is associated with a reduced risk of GIB in patients using DAPT after AMI, with an absolute hazard ratio of 0.62 (95% CI: 0.48-0.77) and a 1-year absolute risk difference of 0.44% (95% CI: 0.39-0.48).[5]We also found that PPI prescribing have increased over time, from 11.2% in 2003 to 24.3% in 2014 (P= 0.0001) but was still far below guideline recommendations.

Smoking (OR = 1.494, 95% CI: 1.054-2.118,P=0.024) was found to be an independent risk factor for GIB in the STEMI patients included in this study(Table 3).The parietal cells of the stomach are more prone to hyperplasia and irritation in patients who smoke.The secreted gastric acid inhibits the synthesis of prostaglandins in the gastroduodenal mucosa and reduces the blood flow of the gastric mucosa, ultimately disrupting the protective barrier of the digestive tract.[10]In addition, smoke inhalation inhibits the activity of NO synthase in the blood and hinders the synthesis of epidermal growth factor in the gastric mucosa.This effect is detrimental to the healing of lesions, such as ulcers.[11]However, the findings of a previous study[12]suggested that smoking status is one of the strongest predictors of GIB, which may be related to the higher infection rate ofHelicobacter pylori(HP) among smokers.HP infection is an independent risk factor for GIB.Notably, this spiralshaped bacterium can express mediators that disrupt the gastric mucosal barrier function, such as protease and phospholipase A.Furthermore, the absorption of gastric and exogenous nutrients is impeded, thereby affecting gastric function.Active, accompanied by HP infection, and synergism with antiplatelet drugs can aggravate digestive tract injury.[13]An earlier study found that the prevalence of Hp in patients with acute myocardial infarction was 20%,and HP-positive status was more common in smokersthan in non-smokers.[14]

It is well known that inflammatory factors represented by CRP play an important role in the occurrence and prognosis of acute myocardium infarction.HP infection is persistent and stimulates local and systemic immune responses, increasing the levels of CRP, interleukin-8, and tumor necrosis factor a, which can be significantly decreased after eradication of HP.[15]However, atherosclerosis and a prethrombotic state can be induced by the increased inflammatory response in the body, which eventually leads to ischemic heart disease and AMI occurrence.[15]An earlier study found that the prevalence of Hp in patients with AMI was 20%.[14]Another investigation suggested that a correlation existed between CRP and HP infection and between CRP and rebleeding in acute non-variceal GIB.[16]In this study, CRP was measured at baseline; we established that CRP levels in patients with GIB were significantly higher than those in patients without GIB.Logistic regression analysis results showed that Hs-CRP (OR = 1.020, 95% CI: 1.001-1.038,P= 0.036) was an independent risk factor for GIB in STEMI patients.

Effective antithrombotic therapy has significantly improved the outcomes in patients with AMI,but at the cost of an increased risk of bleeding.Chronic gastric infection with HP usually leads to GIB.Therefore, it can be speculated that the significant increase of CRP in patients with GIB may reflect the presence of HP infection.However, no research evidence on the relationship between CRP and HP infection has been published.

Triiodothyronine (T3) levels are independently associated with 30-day and 1-year all-cause mortality in patients with acute myocardial infarction, and low T3 levels are closely associated with poor prognosis in STEMI patients.[17]T3 (OR = 0.513, 95% CI 0.297-0.889,P= 0.017) was found to be an independent risk factor for GIB in STEMI patients.The results of a previous study suggested that lower serum FT3 and higher CRP levels were associated with a higher risk of death in AMI patients, with a 2.5-fold increase in the risk of death.[18]In animal models of acute myocardial infarction, supplemental T3 therapy modulated the inflammatory process.[19]Moreover, low T3 level was associated with high CRP,which may reflect a more intense inflammatory response in STEMI patients leading to poor prognosis.[20]An excessive inflammatory response puts the body in a state of stress, leading to ischemia of the gastric mucosa due to vasospasm, decreased mucus secretion, lack of prostaglandin synthesis, mucosal barrier destruction, mucosal erosion, bleeding,and even ulcer development.In addition, a crosssectional study including 948 participants aged from 30 to 85 years based on the 1999-2000 US National Health and Nutrition Examination Survey found that high thyroid stimulating hormone (TSH) levels were associated with HP infection.[20]However, this study does not find that high TSH level increases the risk of GIB in STEMI patients.

In conclusion, we established that the incidence of in-hospital GIB in STEMI patients was 11.63%, and the in-hospital mortality was 3.2%.Importantly, hs-CRP, triiodothyronine, and smoking were identified as independent risk factors for in-hospital GIB in STEMI patients.However, this study has limitations, such as its retrospective design, the lack of HP infection detection, and a thyroid function test performed after interventional treatment, which cannot completely exclude the influence of iodinated contrast media on the detection.