a b s t r a c t

Background: Research showed that the mortality of Upper gastrointestinal hemorrhage (UGH) complicated with acute coronary syndrome (ACS) was higher than single UGH in elderly patients. This study aimed to determine the risk factors that associated with an increased risk of ACS occurrence after UGH.

Methods: A population-based Nested case-control study was conducted analyzing the hospital information sys- tem database of Shengjing Hospital of China Medical University from September 1, 2009 to December 31, 2014. We included 3217 elderly patients who experienced a UGH, among which 152 cases were identified and matched 604 selected controls. Multivariate conditional logistic regression models were used to characterize risk factors associated with ACS occurrence and death after UGH.

Results: Diabetes (odds ratio (OR) = 1.84, 95% confidence interval (CI) 1.13-2.71, P = 0.039), smoking (OR = 1.87, 95% CI 1.19-2.73, P = 0.028), vasopressin or terlipressin use (OR = 1.51, 95% CI 1.02-2.14, P = 0.043),

liver cirrhosis (OR = 2.43, 95% CI 1.45-4.38, P = 0.013), hemoglobin level (OR = 2.36, 95% CI 1.65-3.79, P = 0.014) and history of ACS (OR = 1.98, 95% CI 1.13-3.49, P = 0.017) increased risk of ACS incidence in elderly pa- tients with UGH. Moreover, diabetes (OR = 2.14, 95% CI 1.15-4.21, P = 0.041), smoking (OR = 2.93, 95% CI 1.17-5.31, P = 0.043) and Hemoglobin levels (OR = 1.95, 95% CI 1.24-3.16, P = 0.038) were independent var- iables for the mortality underwent UGH with ACS in elderly patients.

Conclusions: History of diabetes, vasopressin or terlipressin use, smoking, liver cirrhosis, hemoglobin level and history of ACS are risk factors to develop ACS in elderly patients with UGH. Importantly, diabetes, smoking and lower hemoglobin level are key variables for mortality.

(C) 2018

Introduction

Acute upper gastrointestinal hemorrhage (UGH) is a very com- mon diagnosis in emergency department. The incidence is reported to be 500/100,000 per year in those aged over 75 years, compared with that of around 48-160/100,000 per year in the general popula- tion in western countries, and lead to 300,000 admission every year [1-7], costing nearly five billion dollars annually to the health system in USA [8]. In most cases, it is caused by nonvariceal factors, and the mortality is up to 3-14% [9]. Some patients with UGH may suffer

Abbreviations: UGH, upper gastrointestinal hemorrhage; ACS, acute coronary syndrome; OR, odds ratio; CI, confidence interval; HIS, hospital information system; STEMI, ST-segment elevation myocardial infarction; NSTEMI, non-ST-segment elevation myocardial infarction; UA, unstable angina; NSAID, nonsteroidal anti-inflammatory drug.

* Corresponding authors.

E-mail addresses: [email protected], (T. Li), [email protected] (Q. Du).

¹ The authors contributed equally to the work.

from dyspnea and chest pain, which indicate that they may develop acute coronary syndrome (ACS). It is reported that cardiovascular disease or its complications is often the cause of death among elderly patients with UGH. The mortality of UGH with ACS is up to 62% [10], much higher than the 20% mortality of patients with UGH alone [11]. But, the occurrence of ACS after UGH is frequently ignored, because the signs and symptoms may be covered up by severe UGH [1]. Until now, the risk factors, incidence, and outcomes associated with ACS in UGH patients have not been clear yet. Given this, increased awareness of potential risk factors for ACS after UGH is imperative to reduce mortality.

So, in this study, we carried out study on ACS in elderly patients with UGH presenting to the Emergency Department in Shengjing Hospital of China Medical University from September 2009 to December 2014. We hypothesized that ACS may develop in patients with UGH, and we ex- amined the clinical variables associated with concurrent ACS and UGH. We aimed to identify risk factors for the development of ACS in el- derly patients with UGH.

https://doi.org/10.1016/j.ajem.2018.06.055

0735-6757/(C) 2018

Methods

Data source

We conducted a population-based retrospective nested case-control study to assess the association between risk factors and ACS occurrence in patients with UGH. The UGH patients were identified from the hospi- tal information system (HIS) database of Shengjing Hospital of China Medical University. Shengjing Hospital is the largest hospital located in northeast in China, Liaoning Province. The HIS system is the electronic medical recording system used in both the inpatient and outpatient set- tings at Shengjing Hospital. The HIS system provided detail of each pa- tient, including date of admission and discharge, general information, present history, past history, personal history, medical procedure dur- ing the hospital stay, medical procedure, examination results, primary and secondary diagnoses, etc.

Study population and design

We identified patients presenting to the Emergency Department from September 1, 2009 to December 31, 2014, with a primary diagno- sis of UGH (ICD-10 code of K92.208, gastrointestinal hemorrhage; K92.206, gastric hemorrhage; K92.204, upper gastrointestinal hemor- rhage; K92.101, melena; K92.001, hematemesis). A patient developed ACS after diagnosis of UGH was defined as case, no ACS was selected as control. By using an incidence density sampling approach, each case was matched by age (+-5 years), sex and entry date (+-30 days) with up to 4 randomly selected controls.

We assessed the mortality of patients at one month post-admission. They have detailed records about death causes in our HIS system. If the patients had been discharged from our hospital, we assessed their out- comes by telephone call performed by an Emergency Physician. We identified the causes from patients’ death proof, in which death cause is recorded by doctors. If the patient died at home, with no death proof, the cause is identified as unknown.

The protocol used in this study was approved by the ethics commit-

tee of Shengjing Hospital of the China Medical University, and all sub- jects gave their written informed consent to participate.

Inclusion and exclusion criteria

Upper gastrointestinal bleeding was defined as symptoms of tarry stools, melena, coffee-ground vomitus, hematemesis, or anemia suspected to be related to UGI bleeding. Endoscopy within 24 h of pre- sentation following successful resuscitation is carried out for most pa- tients with UGH. Patients with variceal bleeding or high risk stigmata were examined by endoscopy in 12 h.

ACS was defined as ST-segment elevation myocardial infarction , Non-ST-segment elevation myocardial infarction (NSTEMI), or unstable angina . We defined elderly patients as >=65 years of age. Exclusion criteria were hematological diseases with coagulation dis- orders, history of myocardial bridge, serious infection, severe neurolog- ical disease, trauma, pregnancy, and recent myocardial infarction (within 6 weeks of emergency department admission). For patients who died shortly after admission before evaluations for ACS was

excluded.

Risk factors

Electrocardiography and cardiac enzymes, including creatine kinase (CK), CK-MB, and TnI, were checked after admission. Liver cir- rhosis was previously diagnosed in hospital when a typical change in abdomen by ultrasound or CT or MRI, in addition to a hepatitis history or evidence of hepatic decompensation. We evaluated the five tradi- tional cardiac risk factors (diabetes, hypertension, age, hypercholester- olemia and smoking), vasopressin or terlipressin use, alcohol use,

drugs used and cause of bleeding, as well as other risk factors (age, male gender, obesity). By early observation, patients seldom had ACS if hemoglobin N90 g/L, so in this study, we selected patients presented with initial hemoglobin<=90 g/L. We classified anemia by hemoglobin level into three levels: <=30 g/L as very Severe anemia, <=60 g/L as severe anemia, <=90 g/L as moderate anemia. Previous UGH and myocardial in- farction history, underlying malignancy and drug use including aspirin, coumadin, nonsteroidal anti-inflammatory drug (NSAID) and steroid, and history of tobacco and alcohol use were recorded in our system.

Statistical methods

All statistical analyses were performed using SAS 9.2 software. Data was expressed as mean +- SD or median (25th-75th percentiles) or as proportion. The association of variables with ACS was expressed as odds ratio (OR) with 95% confidence interval (CI), quantified by using a conditional logistic analysis. The sample size was calculated according to method of 1:4 matching Case-control studies, with OR = 2, exposure rate in control group = 0.2, rate of error type I = 0.05 and rate of error type II = 0.10. Variables with P b 0.1 in univariate analysis were entered into multiple logistic regression analysis using a backward algorithm. P b 0.05 was considered statistically significance in the multivariate con- ditional logistic regression.

Results

A total of 3217 patients who presented with UGH from September 1, 2009 to December 31, 2014 were identified. 203 patients were excluded according to exclusion criteria, leaving 3014 patients as samples. Among them, 152 were diagnosed as elderly patients with ACS as experiment group. 604 without ACS were selected randomly as controls.

All the clinical characteristics were compared in Table 1. Patients in the cases group had higher BMI (odds ratio = 1.23, 95% confidence in- terval 1.06-1.81, P = 0.043), higher incidence of diabetes (OR = 2.68, 95% CI 1.72-4.16, P b 0.001), hypertension (OR = 1.38, 95% CI

0.95-2.01, P = 0.093), hypercholesterolemia (OR = 1.67, 95% CI

0.94-2.98, P = 0.083), smoking (OR = 1.90, 95% CI 1.26-2.86, P =

0.002), vasopressin or terlipressin use (OR = 1.99, 95% CI 1.22-3.27, P

= 0.006), alcohol (OR = 1.59, 95% CI 1.10-2.29, P = 0.014), higher in-

cidence of liver cirrhosis (OR = 5.39, 95% CI 3.50-8.30, P b 0.001), lower blood hemoglobin level after bleeding (OR = 1.72, 95% CI 1.32-2.25, P b 0.001) and history of ACS (OR = 1.91, 95% CI 1.08-3.39, P = 0.027).

Given these findings, we selected BMI, hemorrhage volume, incidence of diabetes, hypertension, hypercholesterolemia, vasopressin or terlipressin use, alcohol, liver cirrhosis, blood hemoglobin level, and his- tory of ACS as variables for multivariate conditional logistic regression in the subsequent evaluation.

In the multivariate conditional logistic regression analysis, incidence of diabetes (OR = 1.84, 95% CI 1.13-2.71, P = 0.039), smoking (OR = 1.87, 95% CI 1.19-2.73, P = 0.028),vasopressin or terlipressin use (OR

= 1.51, 95% CI 1.02-2.14, P = 0.043), liver cirrhosis (OR = 2.43, 95%

CI 1.45-4.38, P = 0.013), hemoglobin level (OR = 2.36, 95% CI

1.65-3.79, P = 0.014) and history of ACS (OR = 1.98, 95% CI

1.13-3.49, P = 0.017) remained significant associated with a higher odds of developing ACS, while BMI (OR = 1.12, 95% CI 0.84-1.43, P = 0.072), alcohol (OR = 1.52, 95% CI 0.73-1.94, P = 0.079), hypertension (OR = 1.16, 95% CI 0.87-1.84, P = 0.134) and hypercholesterolemia (OR = 1.43, 95% CI 0.83-2.15, P = 0.091) were not found to be signifi- cant risk factors (Fig. 1).

In the subsequent step, we analyzed the association of variables with mortality of UGH complicated by ACS in elderly patients. We assessed hospital outcomes of admitted patients at one month post-admission. If the patients had been discharged from the hospital, we assessed their outcomes by telephone call performed by an Emergency Physician. Causes of death were identified in 24 patients including Cardiovascular death (45.8%, n = 11), GI bleeding (25%, n = 6), sepsis (16.7%, n =

Table 1

Clinical characteristics of UGH in elderly patients.

Variables	Cases (n = 152)	Controls (n = 604)	OR (95% CI) Pa
Age (yrs)	72.34 +- 11.19	71.73 +- 9.88	1.06 (0.65-1.34)	0.84
Male sex % (n)	59.8 (91)	57.5 (347)	1.11 (0.77-1.59)	0.59
BMI	25.2 +- 2.6	24.5 +- 1.9	1.23 (1.06-1.81)	0.043
Diabetes % (n)	25.7 (39)	11.4 (69)	2.68 (1.72-4.16)	b0.001
Hypertension % (n)	36.2 (55)	29.1 (176)	1.38 (0.95-2.01)	0.093
Hypercholesterolemia % (n)	11.8 (18)	7.5 (45)	1.67 (0.94-2.98)	0.083
Smoking % (n)	28.3 (43)	17.2 (104)	1.90 (1.26-2.86)	0.002
Vasopressin or terlipressin use % (n)	17.8 (27)	9.8 (59)	1.99 (1.22-3.27)	0.006
Alcohol % (n)	40.8 (62)	30.3 (183)	1.59 (1.10-2.29)	0.014
Liver cirrhosis % (n)	35.5 (54)	9.3 (56)	5.39 (3.50-8.30)	b0.001
Hemoglobin (g/L) <=30% (n)	61.8 (94)	42.1 (254)	1.72 (1.32-2.25)	b0.001
<=60% (n)	27.0 (41)	39.2 (237)
<=90% (n)	11.2 (17)	18.7 (113)
History of ACS % (n)	12.5 (19)	7.0 (42)	1.91 (1.08-3.39)	0.027
History of drugs usage Anti-coagulation drugb % (n)	8.6 (13)	8.9 (54)	0.95 (0.51-1.80)	0.88
Antiplatelets drugc % (n)	51.3 (78)	52.3 (316)	0.96 (0.67-1.37)	0.83
NSAIDsd % (n)	42.8 (65)	35.6 (215)	1.35 (0.94-1.94)	0.103
Steroid % (n)	5.3 (8)	3.1 (19)	1.71 (0.73-3.99)	0.214
proton pump inhibitor % (n)	75 (114)	76.7 (463)	0.91 (0.61-1.38)	0.67
Cause of bleeding Peptic ulcere % (n)	45.4 (69)	51.5 (311)	0.88 (0.73-1.05)	0.149
Varices % (n)	30.9 (47)	25.3 (153)
Cancer % (n)	11.2 (17)	16.7 (101)
Othersf % (n)	12.5 (19)	6.5 (39)

UGH, upper gastrointestinal hemorrhage; ACS, acute coronary syndrome; NSAID, nonsteroidal anti-inflammatory drug.

^a Significance was obtained by conditional logistic regression.

^b Warfarin and apixaban, use were categorized as use of anti-coagulations.

^c Aspirin and clopidogrel use were categorized as use of antiplatelets.

^d Acetaminophen, ibuprofen, celecoxib, etc., use were categorized as use of NSAIDs.

^e Including esophageal, gastric, and Duodenal ulcers.

^f Including esophagitis, acute gastric mucosal lesion, and gastric erosion.

4), and others (12.5%, n = 3). Cause of cardiovascular death included myocardial infarction, malignant arrhythmia and heart failure, and mis- cellaneous. Of the three patients who died from other causes, one pa- tient died from Hepatic encephalopathy, and two died from unknown causes. After adjusting for the confounding variables of age, other pre- dictors of in-hospital mortality was shown in Table 2. Deceased patients with UGH and ACS had higher incidence of diabetes (OR = 2.73, 95% CI 1.09-6.81, P = 0.032) and liver cirrhosis (OR = 2.95, 95% CI 0.98-4.87, P

= 0.064), smoking (OR = 3.24, 95% CI 1.25-8.41, P = 0.016), lower he-

moglobin levels (OR = 2.16, 95% CI 0.94-4.95, P = 0.068), and history of ACS (OR = 2.95, 95% CI 0.99-8.75, P = 0.051). Thus, in the subsequent step, we selected diabetes, liver cirrhosis, hemoglobin levels and history of ACS as variables for multivariate conditional logistic regression.

In the subsequent multivariate conditional logistic regression analy- sis, diabetes (OR = 2.14, 95% CI 1.15-4.21, P = 0.041), smoking (OR = 2.93, 95% CI 1.17-5.31, P = 0.043) and lower hemoglobin levels (OR =

Fig. 1. Risk factors for ACS in elderly UGH patients. Multivariate conditional logistic regression analysis demonstrated diabetes, vasopressin or terlipressin use, smoking, liver cirrhosis, hemoglobin level and history of ACS as independent variables for ACS in elderly UGH patients. OR, odds ratio; CI, confidence interval.

1.95, 95% CI 1.24-3.16, P = 0.038) were identified as independent var- iables for the mortality in elderly patients with UGH complicated by ACS, while liver cirrhosis (OR = 2.02, 95% CI 0.83-5.19, P = 0.076), and history of ACS (OR = 2.53, 95% CI 0.81-5.34, P = 0.059) were not statistically significant (Fig. 2).

Discussion

ACS is a serious complication of UGH. It is reported that the morbid- ity of severe UGH with ACS is 5-10% in intensive care units [12], and it is 0.94% in patients with UGH outside intensive care units [13]. The mech- anism of ACS in UGH patient is thought to be related to hypovolemia, vasoconstriction and low myocardial perfusion, particularly in patients with underlying coronary disease, diabetes mellitus, smoking history, inadequate fluid supply and gastrointestinal bleeding [14]. The symp- toms and signs of ACS can be disguised by those of UGH such as hypo- tension, abdominal pain, nausea, vomiting, and sudden dysphoria, particularly in elderly patients. Elderly patients often exhibit subtle symptoms of ACS, which can increase the incidence of misdiagnosis.

Our observational nested case-control study indicated that patients with cirrhosis were at high risk to develop myocardial injury during acute UGH. We analyzed the use of vasopressin and terlipressin, which is common in treatment of variceal bleeding. This study sug- gested that vasopressin and terlipressin may cause coronary vasocon- striction as a side effect. The results concur with the findings of previous studies [12, 15]. This finding emphasizes the importance of close attention to possible ACS while treating UGH in elderly patients with cirrhosis.

In this study, we analyzed the presence of traditional cardiac risk fac- tors as an independent variable in the development of ACS in UGH pa- tients. These Traditional risk factors include diabetes, hypertension, elderly age, hypercholesterolemia, and Tobacco use. We found that UGH patients with diabetes and smoking had a higher incidence of

Table 2

Variables of mortality underwent UGH with ACS in elderly patients.

Variables	Death (n = 24)	Non-death (n = 128)	OR (95% CI) Pa
Age (yrs)	81.4 +- 12.3	74.8 +- 9.4	-	-
Male sex n (%)	62.5 (15)	59.4 (76)	1.14 (0.46-2.80)	0.775
BMI	25.4 +- 2.3	25.5 +- 1.7	1.57 (0.85-2.16)	0.24
Diabetes % (n)	41.7 (10)	22.66 (29)	2.73 (1.09-6.81)	0.032
Hypertension % (n)	58.3 (14)	47.7 (61)	1.54 (0.64-3.72)	0.34
Hypercholesterolemia % (n)	16.7 (4)	10.9 (14)	1.63 (0.49-5.45)	0.43
Smoking % (n)	37.5 (9)	15.6 (20)	3.24 (1.25-8.41)	0.016
Vasopressin or terlipressin use % (n)	20.8 (5)	17.2 (22)	1.27 (0.43-3.76)	0.67
Alcohol n (%)	54.2 (13)	38.3 (49)	1.91 (0.79-4.59)	0.15
Liver cirrhosis n (%)	45.8 (11)	30.5 (38)	2.95 (0.98-4.87)	0.064
Hemoglobin (g/L) <=30	75.0 (18)	59.4 (76)	2.16 (0.94-4.95)	0.068
<=60	25.0 (6)	27.3 (35)
<=90	0 (0)	13.3 (17)
History of ACS % (n)	25 (6)	10.15 (13)	2.95 (0.99-8.75)	0.051
History of drugs usage Anti-coagulation druga n (%)	12.5 (3)	7.8 (10)	1.69 (0.43-6.64)	0.46
Antiplatelets drugb n (%)	58.3 (14)	50.0 (64)	1.40 (0.58-3.38)	0.46
NSAIDsc n (%)	50.0 (12)	41.4 (53)	1.42 (0.59-3.39)	0.44
Steroid n (%)	8.3 (2)	4.7 (6)	1.85 (0.35-9.76)	0.47
Cause of bleeding Peptic ulcerd	33.3 (8)	47.7 (61)	0.95 (0.62-1.44)	0.79
Varies	45.8 (11)	28.1 (36)
Cancer	12.5 (3)	10.9 (14)
Otherse	8.3 (2)	13.3 (17)

UGH, upper gastrointestinal hemorrhage; ACS, acute coronary syndrome; NSAID, nonsteroidal anti-inflammatory drug.

^a Warfarin and apixaban, use were categorized as use of anti-coagulations.

^b Aspirin and clopidogrel use were categorized as use of antiplatelets.

^c Acetaminophen, ibuprofen, celecoxib, etc., use were categorized as use of NSAIDs.

^d Including esophageal, gastric, and duodenal ulcers.

^e Including esophagitis, acute gastric mucosal lesion, and gastric erosion.

ACS. Moreover, diabetes and smoking are also independent variables for mortality. Thus, elderly patients with diabetes and smoking have higher risk of ACS after UGH, and the mortality is much higher than those with- out these underlying risk factors.

In our study, we identified hemoglobin level as a risk factor for de- velopment of ACS in UGH patients. In our subsequent analysis, we also confirmed it as an individual predictor of mortality in elderly UGH pa- tients with ACS. Thus, hemoglobin level is an important index to moni- tor ACS of UGH patients. The importance of closely monitoring hemoglobin is in accordance with the present treatment recommenda- tions [16]. In these recommendations, blood transfusion is advised when the hemoglobin level is b70 g/L in UGH patients. Transfusion can rectify anemia and supplement blood volume, improve coronary va- soconstriction, and improve systemic and local Oxygen exchange. It is also reported that blood transfusion can decrease the incidence of ACS in elderly patients with coronary disease, diabetes mellitus and tobacco use [14].

The treatment of UGH with ACS is complex, particularly regarding the use of medications that affected clotting. The usage of hemostatic drugs for UGH can increase the morbidity of ACS, while anticoagulants

Fig. 2. risk factors for mortality in elderly patients with UGH complicated by ACS. By multivariate conditional logistic regression analysis, diabetes, smoking and hemoglobin level are independent variables for mortality of UGH with ACS. OR, odds ratio; CI, confidence interval.

and Antiplatelet agents used in treating ACS can worsen gastrointestinal bleeding. In our study, we observed that pharmacologic hemostasis in- creased the incidence of ACS in UGH patients. Despite this, usage of he- mostasis, anti-coagulations and antiplatelets drugs did not increase the mortality. So, how to utilization of medications that affected clotting is an important problem in treatment. At present, it is recommended that aspirin should be taken as soon as UGH is controlled when cardio- vascular risks are greater than gastrointestinal ones [17]. It is generally advised that aspirin or clopidogrel should be stopped for 24 h after gas- trointestinal bleeding. After assessing the gastrointestinal and bleeding conditions, clopidogrel can be restarted in 1-2 days, and aspirin in 1-2 weeks [18, 19]. Until now, there is no clear-cut recommendations on hemostasis in patients with UGH complicated by ACS.

There are some limitations in our investigation. We only included

152 cases and 604 controls from one single hospital. A large scale, mul- ticenter investigation is needed in the future. But, low development de- gree of shared electric hospital information in China limits the gain of data. We need a nationally shared database, locally at least, to collect more data. Meanwhile, our investigation supplied important informa- tion to alert ACS in possible elderly UGH patients, especially with diabe- tes, smoking and lower hemoglobin level. Further investigations on interventions to prevent the development of ACS after UGH are needed.

Conclusions

In conclusion, the mortality of UGH with ACS is much higher than in UGH or ACS alone. Our study showed that elderly patient with diabetes, more vasopressin or terlipressin use, smoking, liver cirrhosis, low he- moglobin level and history of ACS are at increased risk of developing ACS during an episode of UGH. Moreover, diabetes, smoking and hemoglobin level are key predictors of mortality in these patients. These findings suggest that it is imperative to carefully monitor electro- cardiographic findings and cardiac enzymes in elderly UGH patients to recognize ACS, particularly in those with underlying risk factors, and thus reduce mortality.

Grants

This work was supported by the National Natural Science Founda- tion of China (Nos. 81772056 and 81500628) and grant of Education Department of Liaoning Province (No. L2015568).

References

1. Gonzalez-Gonzalez JA, Monreal-Robles R, Garcia-Compean D, Paz-Delgadillo J, Wah- Suarez M, Maldonado-Garza HJ. Nonvariceal upper gastrointestinal bleeding in el- derly people: clinical outcomes and prognostic factors. J Dig Dis 2017;18:212-21. https://doi.org/10.1111/1751-2980.12459.
2. Jairath V, Martel M, Logan RF, Barkun AN. Why do mortality rates for nonvariceal upper gastrointestinal bleeding differ around the world? A systematic review of co- hort studies. Can J Gastroenterol 2012;26:537-43.
3. Laine L, Yang H, Chang SC, Datto C. Trends for incidence of hospitalization and death due to GI complications in the United States from 2001 to 2009. Am J Gastroenterol 2012;107:1190-5. https://doi.org/10.1038/ajg.2012.168 (quiz 6).
4. Lassen A, Hallas J, Schaffalitzky de Muckadell OB. Complicated and uncomplicated peptic ulcers in a Danish county 1993-2002: a population-based cohort study. Am J Gastroenterol 2006;101:945-53. https://doi.org/10.1111/j.1572-0241.2006.00518. x.
5. Rockall TA, Logan RF, Devlin HB, Northfield TC. Incidence of and mortality from acute upper gastrointestinal haemorrhage in the United Kingdom. Steering Committee and members of the National Audit of Acute Upper Gastrointestinal Haemorrhage. BMJ 1995;311:222-6.
6. Targownik LE, Nabalamba A. Trends in management and outcomes of acute nonvariceal upper gastrointestinal bleeding: 1993-2003. Clin Gastroenterol Hepatol 2006;4:1459-66. https://doi.org/10.1016/j.cgh.2006.08.018.
7. van Leerdam ME, Vreeburg EM, Rauws EA, Geraedts AA, Tijssen JG, Reitsma JB, Tytgat GN. Acute upper GI bleeding: did anything change? Time trend analysis of in- cidence and outcome of acute upper GI bleeding between 1993/1994 and 2000. Am J Gastroenterol 2003;98:1494-9. https://doi.org/10.1111/j.1572-0241.2003.07517.x.
8. Peery AF, Crockett SD, Barritt AS, Dellon ES, Eluri S, Gangarosa LM, Jensen ET, Lund JL, Pasricha S, Runge T, Schmidt M, Shaheen NJ, Sandler RS. Burden of gastrointestinal, liver, and pancreatic diseases in the United States. Gastroenterology 2015;149: 1731-41. https://doi.org/10.1053/j.gastro.2015.08.045 e3.
9. X Q HN, Hu X. Analysis of reasons and clinical manifestation of upper gastrointesti- nal hemorrhage in 1520 cases. Clin J Gastrointest Dis 2011;23:12-5.
10. Phan SH. Biology of fibroblasts and myofibroblasts. Proc Am Thorac Soc 2008;5: 334-7. https://doi.org/10.1513/pats.200708-146DR.
11. Blocksom JM, Tokioka S, Sugawa C. Current therapy for nonvariceal upper gastroin- testinal bleeding. Surg Endosc 2004;18:186-92. https://doi.org/10.1007/s00464- 003-8155-4.
12. Bhatti N, Amoateng-Adjepong Y, Qamar A, Manthous CA. Myocardial infarction in critically ill patients presenting with gastrointestinal hemorrhage: retrospective analysis of risks and outcomes. Chest 1998;114:1137-42.
13. Cappell MS. A study of the syndrome of simultaneous acute upper gastrointestinal bleeding and myocardial infarction in 36 patients. Am J Gastroenterol 1995;90: 1444-9.
14. Lee CT, Huang SP, Cheng TY, Chiang TH, Tai CM, Su WC, Huang CH, Lin JT, Wang HP. Factors associated with myocardial infarction after Emergency endoscopy for upper gastrointestinal bleeding in high-risk patients: a prospective observational study. Am J Emerg Med 2007;25:49-52. https://doi.org/10.1016/j.ajem.2006.04.013.
15. Wu IC, Yu FJ, Chou JJ, Lin TJ, Chen HW, Lee CS, Wu DC. predictive risk factors for upper gastrointestinal bleeding with simultaneous myocardial injury. Kaohsiung J Med Sci 2007;23:8-16. https://doi.org/10.1016/S1607-551X(09)70368-7.
16. Barkun AN, Bardou M, Kuipers EJ, Sung J, Hunt RH, Martel M, Sinclair P, International Consensus Upper Gastrointestinal Bleeding Conference G. International consensus recommendations on the management of patients with nonvariceal upper gastroin- testinal bleeding. Ann Intern Med 2010;152:101-13. https://doi.org/10.7326/0003- 4819-152-2-201001190-00009.
17. Sung JJ, Chan FK, Chen M, Ching JY, Ho KY, Kachintorn U, Kim N, Lau JY, Menon J, Rani AA, Reddy N, Sollano J, Sugano K, Tsoi KK, Wu CY, Yeomans N, Vakil N, Goh KL, Asia-Pacific Working G. Asia-Pacific Working Group consensus on non-variceal upper gastrointestinal bleeding. Gut 2011;60:1170-7. https://doi.org/10.1136/gut. 2010.230292.
18. Henriksen PA, Palmer K, Boon NA. Management of upper gastrointestinal haemor- rhage complicating dual anti-platelet therapy. QJM 2008;101:261-7. https://doi. org/10.1093/qjmed/hcm148.
19. Luo H, Qin T. Strategy for prevention and treatment of acute myocardial infarction with gastrointestinal bleeding. J China-Jpn Friendship Hosp 2013;27:44-7. https:// doi.org/10.3969/j.issn.1001-0025.2-13.01.013.