Evaluation of Upper Gastrointestinal Bleeding and Its Relationship with Death Incidence

doi:10.21203/rs.3.rs-2125349/v1

Download PDF

Research Article

Evaluation of Upper Gastrointestinal Bleeding and Its Relationship with Death Incidence

https://doi.org/10.21203/rs.3.rs-2125349/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background:

Upper gastrointestinal bleeding (UGIB) is a life-threatening medical emergency characterized by bleeding from the esophagus, stomach, or duodenum. This study aims to analyze the factors associated with the occurrence of death in patients.

Methods:

This retrospective study was conducted in the period between 2021–2022, the hospital’s consent was obtained to view patients’ data, samples were collected from patient records in the Gastrointestinal Division of the hospital. The study included 199 patients admitted to the Gastrointestinal Division who were diagnosed with gastrointestinal bleeding at Al-Mujtahid Hospital in Damascus, the exclusion criteria included both archival errors, and cases with incomplete data that cannot be studied.

Results:

Out of the 199 patients, 68.8% were male and 31.2% were female. The average age of patients was 57.4 years, melena was the most prominent clinical manifestation of bleeding, with 68.2% of patients, followed by Coffee ground vomiting by 34.2%, 30.2% with a history of unexplained hypotension, and 26.1% with a history of haematemesis. The majority used the Rockall warning system with a percentage of 98.5%. The majority had concomitant heart disease 31.2%. Considering the drugs used, aspirin, corticosteroids, and clopidogrel were the top results, 39.9% of patients had a history of heart, 96.5% of patients were conscious, and endoscopy revealed active bleeding in 34.2%. 69.3% had gastric ulcers, 25.1% had esophageal varices, and the majority had a Forrest Ia rating of 32.2% and a grade III of 30.2%. 37.2% needed endoscopic intervention, 19.1% had banding, 11.6% had thermal coagulation, and 10.1% had laparotomy. Death occurred in 6.5% of patients, while 83.4% needed a surgery. Our results found a statistically significant relationship between the incidence of death and each of the advanced age, cardiac antecedents, haematemesis, Coffee ground vomiting, and esophageal varices.

Conclusion:

Our findings reveal the association of various factors with the incidence of death in cases of UGIB reviewed at Damascus Hospital.These results are significant but may not be generalizable, so we recommend conducting more comprehensive studies.

Gastroenterology & Hepatology

Hospital Medicine

Gastrointestinal bleeding

death incidence

retrospective study

Syria

Upper gastrointestinal bleeding (UGIB) is one of the most frequent emergency cases that require hospital admission. Moreover, affected patients have a high risk of mortality in which some cases require admission to the intensive care unit [1] [2]. Gastrointestinal bleeding is classified into upper gastrointestinal bleeding (UGIB) and lower gastrointestinal bleeding (LGIB). Gastrointestinal bleeding that occurs above the level of the ligament of Treitz, the suspensory muscle of the duodenum, is called UGIB. However, if bleeding occurs below the level of this ligament, then it is referred to as LGIB [3]. An UGIB source should be considered when hematochezia presents with signs and symptoms of hemodynamic compromise. The majority of admissions for UGIB are peptic ulcer disease, both gastric and duodenal. Other causes of UGIB include gastritis, esophagitis, duodenitis, varices, upper gastrointestinal malignancies, Mallory-Weiss tear, and vascular abnormalities [4]. The usual symptoms are hematemesis, coffee-ground emesis, hematochezia, and melena [2]. UGIB represents 75% of all acute gastrointestinal bleeding cases, and its annual incidence is approximately 80 to 150 per 100,000 per year [5]. In the past few decades, many improvements have been made in the management of UGIB. However, studies showed that the mortality rate remains constant since 1970s averaging between 5–10% despite the advances in the medical field. Clinical risk factors for mortality are age, comorbidity, tachycardia, and low systolic blood pressure [6]. Upper endoscopy is both a diagnostic and therapeutic tool used in the management of UGIB. Early endoscopy allows early diagnosis, provides the opportunity for endoscopic management, and allows early discharge of patients with low-risk findings [7] This study aims to detect and analyze the risk factors for UGIB among admitted patients in the gastrointestinal department at Damascus Hospital. The risk factor assessment can also help detect patients with a risk of gastrointestinal bleeding, enabling health care providers to predict who needs urgent endoscopy and who can be treated with medication or can be safely discharged.

Study design and material:

This retrospective study included patient records in the Gastrointestinal Department in Damascus Hospital in Damascus, Syria, from 2021 until 2022. The purpose and the nature of the research were explained to all participants and were assured of the confidentiality of the information. During the study period, no economic benefits or other incentives were offered.

Subject:

The study included 199 patients who are admitted to the gastrointestinal department and were diagnosed with UGIB. Exclusion criteria were cases with incomplete data because of medical documentation and archiving misstep.

Data Collection Tool:

the questionnaire used to collect data from patients’ medical records at Damascus Hospital was designed by a specialist physician. Data were collected by the researchers on an electronic form supervised by a specialist physician

Statistical analysis:

Data were evaluated using Statistical Package for the Social Sciences (SPSS) (version 25). where, descriptive statistics for all variable was conducted, and analytical (inferential) statistics were used to present and interpret the results and infer them using chi-square test to reach the goal of the study. Moreover, the adopted statistical significance cut-off point was p < 0.05.

Out of the 199 patients, 68.8% were male and 31.2% were female. The mean age of the patients was 57.4 years (Table 1). About half of the patients were smokers however only 5% of the sample consumed alcoholic drinks. The main symptom in 68.2% of the hospitalized patients was melena, 34.2% was coffee-ground emesis, 30.2% was unexplained hypotension, and 26.1% was hematemesis (Table 2). For most of the patients, the Rockall risk scoring system was used to assess and identify patients at risk of adverse outcomes following acute upper gastrointestinal bleeding accounting for 98.5%. in 9% Forrest risk score was used. The median risk score was 2 out of 7 (Table 3). As for used medications, Aspirin, corticosteroids, and clopidogrel were the most frequently used drugs (Table 4).

96.5% of patients were conscious at the presentation. Where gastrointestinal endoscopy revealed active bleeding in 34.2%. 69.3% had gastric ulcers, and 25.1% had esophageal varices. Most of the patients had a Forrest score Ia accounting for 32.2% and grade III in 30.2%. 37.2% of the patients needed endoscopic intervention, 19.1% required variceal ligation (banding), 11.6% required thermal therapy, and 10.1% had laparotomy. Death occurred in 6.5% of patients, while 83.4% needed surgery (Table 5). A statistically significant relationship was found

between demographic factors and death rate, where 92.3% of the patients within the age group ≥50 died. Whereas, 68.6% of the patients within the age group ≥50 recovered (Table 6). As for the relation between death incidence and associated factors with UGIB, a significant statistical relationship was found between death incidence and hematemesis, where about 53.8% of dead cases had experienced hematemesis (p=0.019). Moreover, a higher death incidence was found in patients who suffered from coffee-ground emesis (p=0.001) (Table 7). Diving further into comorbidity factors, a statistically significant result was found between mortality rate and patients with a previous history of cardiac diseases (P=0.034) (Table 8). Moreover, a statistically significant relationship was found between mortality rate and esophageal varices (Table 9).

Table (1): Socio-demographic characteristics (N= 199)
sex
137 (68.8%)	Male
62 (31.2%)	Female
Age
Mean Standard Deviation	Minimum Value	Maximum Value
57.4 (16.4)	19	100
Age Group
39%	<50
61%	³50
Place of residence
87 (43.7%)	Damascus
82 (41.2%)	Rif Damascus
11 (5.5%)	Daraa
3 (1.5%)	Raqqa
2 (1%)	Tartous
4 (2%)	Aleppo
7 (3.5%)	Al-Hasakah
3 (1.5%)	Homs

Table 2: Main symptom
52 (26.1%)	Hematemesis
68 (34.2%)	Coffee-ground emesis
136 (68.3%)	Melena
60 (30.2%)	Unexplained hypotension
39 (19.6%)	Acute abdominal pain
4 (2%)	Hematochezia
2 (1%)	Abdominal pain
16 (8%)	Other

Table (3): Risk score
196 (98.5%)	Rockall
2 (1%)	Child
18 (9%)	Forrest
Total score
Median (standard deviation)	Minimum score	Maximum score
2 (1.9)	0	7

Table (4): Medication:
79 (38.7%)	Aspirin
4 (2%)	Warfarin
23 (11.3%)	Clopidogrel
27 (13.2%)	Corticosteroids
11 (5.4%)	Diclon
12 (5.9%)	Pain killer (brufen, paracetamol)

Table (5): Medical history of upper GI bleeding
General condition
192 (96.5%)	Conscious
2 (1%)	Unconscious
1 (0.5%)	Confusion
Bleeding
68 (34.2%)	Active bleeding
98 (49.2%)	No bleeding
33 (16.6%)	No data
Endoscopic finding
138 (69.3%)	Gastric ulcer
50 (25.1%)	Esophageal varices
9 (4.5%)	Tumors
3 (1.5%)	Abrasion
6 (3%)	Other
Classification Forrest
64 (32.2%)	Ia
15 (7.5%)	Ib
14 (7%)	IIa
8 (4%)	IIb
15 (7.5%)	IIc
60 (30.2%)	III
Total Forrest score
Median (standard deviation)	Minimum score	Maximum score
3.4 (2.2)	1	6
Endoscopic intervention
74 (37.2%)	Yes
109 (54.8%)	NO
Intervention type
23 (11.6%)	Thermal therapy
38 (19.1%)	Endoscopic ligation
13 (6.5%)	Endoscopic clipping
20 (10.1%)	Laparotomy
Outcome
13 (6.5%)	Death
166 (83.4%)	Surgery
19 (9.5%)	Discharge
Death
13 (6.5%)	Yes
185 (93.5%)	No

Table (6): Relation between death and demographic characteristics
		Death		P- Value
		Yes	No
Sex	Male	9 (69.2%)	127 (68.6%)	0.965
	Female	4 (30.8%)	58 (31.4%)
Age	<50	1 (7.7%)	58 (31.4%)	0.007
	≥50	12 (92.3%)	127 (68.6%)

Table (7): relationship between death incidence and variable factors associated with upper gastrointestinal bleeding:
P- Value	Death
P- Value	Yes	No
0.019	7 (53.8%)	45 (24.3%)	Hematemesis	Main symptom
0.001	7 (53.8%)	61 (33%)	Coffee-ground emesis
0.965	9 (69.2%)	127 (68.6%)	Melena
0.508	5 (38.5%)	55 (29.7%)	Unexplained hypotension
0.718	2 (15.4%)	36 (19.5%)	Acute abdominal pain
0.706	0 (0%)	2 (1.1%)	hematochezia
0.269	0 (0%)	16 (8.6%)	Abdominal pain
0.640	13 (100%)	182 (98.9%)	Rockall	Score system
	0 (0%)	2 (1.1%)	Child
	0 (0%)	18 (9.8%)	Forrest
0.870	9 (69.2%)	122 (67%)	> 3	Total score
0.870	4 (30.8%)	60 (33%)	£ 3	Total score
0.913	5 (38.5%)	74 (40%)	Aspirin	Medicines currently used
0.133	1 (7.7%)	3 (1.6%)	Warfarin
0.648	1 (7.7%)	22 (11.9%)	Clopidogrel
0.518	1 (7.7%)	26 (14.1%)	Corticosteroid
0.728	1 (7.7%)	10 (5.4%)	Diclon
0.343	0 (0%)	12 (6.5%)	Other analgesic (brufen, paracetamol)

Table (8): relation between mortality rate and previous medical history:
		Death		P- Value
		yes	No	P- Value
Associated chronic diseases	Cardiac	6 (46.2%)	73 (30.5%)	0.034
	Diabetes mellitus	31 (16.8%)	3 (23.1%)	0.559
	Hematological	5 (2.7%)	0 (0%)	0.548
	Hepatic	1 (7.7%)	31 (16.8%)	0.391
	Previous gastrointestinal bleeding	1 (7.7%)	9 (4.9%)	0.653
Previous blood transfusion	Yes	3 (25%)	47 (27.8%)	0.833
Previous blood transfusion	No	9 (75%)	122 (72.2%)	0.833
Number of unit	Mean (SD)	2 (1)	2.2 (1.4)	0.783
Social habits	Somking	0 (0%)	10 (5.4%)	0.343

Table (9): relationship between mortality and diagnosis and treatment of upper gastrointestinal bleeding:
		Death		P- Value
		Yes	No	P- Value
Endoscopy	Active bleeding	6 (54.5%)	62 (40%)	0.343
Endoscopy	No bleeding	5 (45.5%)	93 (60%)	0.343
Lesion	Gastric ulcer	10 (76.9%)	127 (68.6%)	0.532
	Esophageal varices	2 (15.4%)	7 (3.8%)	0.042
	Tumor	1 (7.7%)	49 (26.5%)	0.132
	Abrasion	0 (0%)	3 (1.6%)	0.644
Forrest classification	Ia	6 (46.2%)	58 (35.6%)	0.696
	Ib	0 (0%)	15 (9.2%)
	IIa	1 (7.7%)	13 (8%)
	IIb	0 (0%)	8 (4.9%)
	IIc	2 (15.4%)	13 (8%)
	III	4 (30.8%)	56 (34.4%)
Endoscopic intervention	Yes	5 (38.5%)	69 (40.8%)	0.867
Endoscopic intervention	No	8 (61.5%)	100 (59.2%)	0.867
Intervention type	Thermal therapy	2 (40%)	21 (23.9%)	0.325
	Band ligation	3 (60%)	35 (39.8%)
	clips	0 (0%)	13 (14.8%)
	Laparotomy	0 (0%)	19 (21.6%)

Acute UGIB is a common cause of death [1]. Therefore, this study describes the mortality pattern among admitted patients in Damascus Hospital from January 2021, until July 2022. This work has many strengths namely, all surveys were filled by trained researchers, which probably minimized the variability in estimating the results. Moreover, questions were designed from several existing surveys used in similar studies. However, the main limitation of the study was the incomplete medical records and inability to contact the patients. Moreover, the accuracy of the results is dependent on the accuracy of database input. Out of the 199 patients, 68.8% were male and 31.2% were female. the median age was 57.4 years. The number of deaths was 13 (6.5%); which is similar to other studies [1] [4]. Our study showed a statistically significant relation between mortality rate and elderly (age greater than 55 years), which is similar to a study done in the United Kingdom where it showed that the most important independent prognostic predictors of mortality at three years were older age; moreover, a similar result was found in a Syrian study [8] [9]. Similar to a Syrian study, our study did not show a statistically significant result that the male gender is a prognostic factor, which is explained by the small size of the sample. Smoking increases the risk of developing gastric ulcers and, to a lesser extent, duodenal ulcers [2]. More than half of the patients in our study were smokers; however, none of them died. The main symptom at the presentation in our study sample was melena (68.3%), followed by coffee-ground emesis (34.2%), which is similar to another Syrian study where out of 155 patients, 70% of them had melena [9]. However, in this study, the second most common symptom was hematemesis, while coffee-ground emesis happened in only 20% of the patients. In our study, about 30.2% of the patients experienced unexplained hypotension, while in another Syrian study less than 5% had unexplained hypotension [9]. These results were also similar to other international studies, where melena and coffee-ground emesis are one of the most commonly reported symptoms of upper gastrointestinal bleeding [1] [6] [8]. Regarding the relationship between symptoms and the incidence of death, a statistically significant relationship was found between hematemesis and coffee-ground emesis and the incidence of death, in contrast to another study where they found a significant relationship between hematochezia and melena, which could be explained by the small sample of our study [1]. Non-steroidal anti-inflammatory drugs (NSAID), including aspirin, are an important risk factor for peptic ulcers and, consequently, upper gastrointestinal bleeding. In our study, 38.7% of the patients used aspirin, and this is similar to other international studies where 53% of duodenal ulcer patients consumed aspirin or other types of NSAID; whereas, 50% of the cases had H. pylori infection [6]. However, the role of H. pylori infection cannot be fully determined in our study because of the lack of biopsies in our sample’s medical records. Comorbid illness is an important prognostic factor, where in our study, cardiac diseases were the most prominent comorbidity, followed by diabetes mellitus and hepatic diseases, respectively. Recent studies found that usually comorbid illnesses are the main cause of death [10], our study showed a statistically significant relation between cardiac diseases (P = 0.034) with the incidence of death, but no other significant results were found with the other comorbidities. Regarding blood transfusion, studies showed transfusing blood more than 6 times is a bad prognostic factor [2]. In our study, 25.1% of the sample required blood transfusion on average 2.2, and the highest was 8 times. Moreover, 34.2% had active bleeding at the presentation, which is considered a bad prognostic factor [2]. However, none of the previous results showed statistically significant results in contrast to another Syrian study which showed a statistically significant relationship between the history of blood transfusion and blood unit and the existence of active bleeding with a bad prognosis [9]. As for endoscopic findings, we found that the most prominent diagnosis was gastric ulcer 138 (69.3%), followed by esophageal varices 50 (25.1%) and tumor 9 (4.5%), respectively. Likewise, a statistically significant relationship was found between the incidence of death and variceal bleeding, which is similar to a study in the United Kingdom. [11]. Nonetheless, no statistically significant relationship was found between the intervention type and the incidence of death.

This retrospective study found that the incidence of death is more likely in elderly patients, the presence of hematemesis or coffee-ground emesis, ruptured esophageal varices, and associated cardiac disease. However, it is not possible to generalize these results in Syria because this study included only the medical records in Damascus Hospital; therefore, we recommend conducting more comprehensive studies, and further studies regarding the treatment options are needed.

Disclosure: The authors report no conflicts of interest in this work

Ethics: Our study was approved by the Institutional Review Board (IRB) at the Syrian Private University (SPU), however, we were not given a reference number. Written informed consent was obtained from all participants.

Hawkey C. Textbook of clinical gastroenterology and hepatology. 2nd ed. Chichester: Wiley-Blackwell; 2012.
El-Oma E. Davidson's principles and practice of medicine. 23rd ed. New York: Elsevier; 2018.
Antunes C, Copelin II EL. Upper Gastrointestinal Bleeding. [Updated 2022 Jul 18]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2022 Jan-. Available from: https://www.ncbi.nlm.nih.gov/books/NBK470300/
Blatchford O, Davidson L, Murray W, Blatchford M, Pell J (1997) Acute upper gastrointestinal haemorrhage in west of Scotland: case ascertainment study. BMJ 315(7107): 510–514.
Sehested TSG, Carlson N, Hansen PW, Gerds TA, Charlot MG, Torp-Pedersen C, Køber L, Gislason GH, Hlatky MA, Fosbøl EL. Reduced risk of gastrointestinal bleeding associated with proton pump inhibitor therapy in patients treated with dual antiplatelet therapy after myocardial infarction. Eur Heart J. 2019 Jun 21;40(24):1963–1970
Feldman M, Friedman L, Brandt L. Sleisenger and Fordtran's gastrointestinal and liver disease. 11th ed. Philadelphia, PA: Elsevier Saunders; 2022.
Hopper AD, Sanders DS. Upper GI bleeding requires prompt investigation. Practitioner. 2011 Jul-Aug;255(1742):15 – 9, 2. PMID: 21932501.
Roberts SE, Button LA, Williams JG (2012) Prognosis following Upper Gastrointestinal Bleeding. PLOS ONE 7(12): e49507. https://doi.org/10.1371/journal.pone.0049507
Bitar S, Moussa M. The risk factors for the recurrent upper gastrointestinal hemorrhage among acute peptic ulcer disease patients in Syria: A prospective cohort study. Annals of Medicine and Surgery [Internet]. 2022 [cited 26 September 2022];74:103252. Available from: https://www.sciencedirect.com/science/article/pii/S2049080122000127?via%3Dihub
Yavorski RT, Wong RK, Maydonovitch C, Battin LS, Furnia A, Amundson DE. Analysis of 3,294 cases of upper gastrointestinal bleeding in military medical facilities. Am J Gastroenterol. 1995 Apr. 90(4):568–73.
Corson JD, Williamson RCN, eds. Surgery. London, UK: Mosby-Year Book; 2001

Download PDF

Version 1

posted

You are reading this latest preprint version

Evaluation of Upper Gastrointestinal Bleeding and Its Relationship with Death Incidence

Status:

Version 1

Abstract

Background:

Methods:

Results:

Conclusion:

Introduction

Methods

Study design and material:

Subject:

Data Collection Tool:

Statistical analysis:

Result

Discussion

Conclusion

Declarations

References

Status:

Version 1