Prediction of the need for emergency endoscopic treatment for upper gastrointestinal bleeding and new score model: A retrospective study

doi:10.21203/rs.3.rs-1667985/v1

PDF

Research Article

Prediction of the need for emergency endoscopic treatment for upper gastrointestinal bleeding and new score model: A retrospective study

https://doi.org/10.21203/rs.3.rs-1667985/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: Gastrointestinal bleeding is one of the major gastrointestinal diseases. In this study, our objective was to compare Glasgow-Blatchford Score (GBS), AIMS65 score, MAP score, Modified GBS, and Iino et al. as outcome measures for upper gastrointestinal bleeding. In addition, we extracted factors associated with hemostatic procedures including endoscopy, and proposed a new robust score model.

Methods: From January 2015 to December 2019, patients with symptoms such as hematemesis who visited the National Hospital Organization Disaster Medical Center and underwent urgent upper endoscopy were retrospectively reviewed. The GBS, AIMS65, MAP score, Modified GBS, and Iino et al. scores were evaluated, and the outcomes were those requiring hemostatic treatments at the subsequent emergency endoscopy. We performed logistic regression analysis of factors related to endoscopic hemostasis and upper gastrointestinal bleeding, and created a new score model to evaluate the prediction of hemostatic treatment and mortality in the new score and the existing scores.

Results: The factors associated with endoscopic treatment were hematemesis, heart rate, HB (Hemoglobin), blood pressure, BUN (Blood urea nitrogen). Based on these predictors and the partial regression coefficients, a new score named HB was generated. HB score was slightly more discriminatory compared to GBS and Modified GBS (area under the receiver operating characteristic curves [AUROC]: 0.73 versus 0.721 and 0.7128, respectively) in predicting hemostatic treatment in emergency endoscopy. The HB score also showed satisfactory prediction accuracy for subsequent deaths (AUROC: 0.6857. P <0.001).

Conclusions: Our new HB score was better performing than the GBS in predicting hemostatic treatment, hemostasis and mortality.

bleeding

blood pressure

endoscopy

hemoglobin

urea

Gastrointestinal bleeding is one of the major gastrointestinal diseases. Upper gastrointestinal bleeding accounts for more than 50% of all gastrointestinal bleeding and related hospitalizations [1], with a mortality rate estimated to be 2-10% [2, 3]. The most common cause of acute upper gastrointestinal bleeding is non-variceal upper gastrointestinal bleeding (NVUGIB).

Urgent endoscopy is performed for suspected gastrointestinal bleeding at night or on holidays, but many cases are experienced in which the bleeding has already stopped spontaneously at the time of endoscopy. Various guidelines indicate the need to stratify patients with gastrointestinal bleeding into high-risk cases requiring immediate treatment before endoscopy and low-risk cases that do not, along with their management before, during, and after endoscopy [4, 5]. If the stratification can reduce the number of urgent endoscopic examinations, it will lead to more efficient medical care and less burden on physicians.

The Glasgow-Blatchford score (hereinafter referred to as "GBS") was reported as an outcome measure for upper gastrointestinal bleeding, which includes comprehensive clinical treatment including blood transfusion, rebleeding, and death in addition to endoscopic treatment [6] (Table 1), and the AIMS65, which measures the risk of death in patients with upper gastrointestinal bleeding, was reported [7] (Table 2). Although the GBS and AIMS65 do not assess the need for endoscopic treatment, they are used to evaluate therapeutic interventions for bleeding, and the usefulness of the GBS has been reported [8-10], because a useful score to predict endoscopic hemostasis is missing.

The GBS has been reported to be useful in the evaluation of therapeutic intervention for bleeding [8-10]. GBS scores range from 0 to 23 points [6]. It has been reported that a score of 7 points or higher is useful as an index for endoscopic treatment [8]. A score of 1 point or lower is low risk and does not require intervention such as endoscopic treatment [10, 11]. The higher the score, the more therapeutic intervention is required, but the evaluation of the point at which therapeutic intervention is required has not been established. Redondo et al. reported that the Glasgow Coma Scale (<15), American Society of Anesthesiologists (ASA) score (>2), pulse (>100 beats/min), albumin (<2.5 mg/dl), systolic blood pressure (<90 mmHg), hemoglobin (Hb We proposed the MAP score, which consists of six indices: pulse (>100 beats/min), albumin (<2.5 mg/dl), systolic blood pressure (<90 mmHg), and hemoglobin (Hb <10 g/dl), and reported that it is highly predictive of therapeutic intervention including endoscopy and mortality in upper gastrointestinal bleeding [12] (Table 3). In order to predict the need for therapeutic intervention including endoscopy in upper gastrointestinal bleeding, the modified GBS consisting of pulse rate, systolic blood pressure, blood urea nitrogen (BUN), and Hb is useful [13, 14] (Table 4).

Iino et al. analyzed 212 Japanese patients to predict the need for endoscopic treatment excluding blood transfusion. Prothrombin time-international normalized ratio (PT-INR), liver disease, heart failure, renal failure, collagen disease, presence of Helicobacter pylori infection, presence of malignancy, antiplatelet agents, anticoagulants, presence of Proton Pump Inhibitor (PPI) or H2 blocker Multivariate analysis was performed to determine whether patients were taking antiplatelet agents, anticoagulants, Proton Pump Inhibitors (PPIs), H2 blockers, corticosteroids, or nonsteroidal anti-inflammatory drugs (NSAIDs). Systolic blood pressure, syncope, hematemesis, Hb, BUN, eGFR, and antiplatelet agents were predictors of the need for therapeutic intervention in gastrointestinal bleeding, and a new scoring model was created (hereafter referred to as the Iino score) [15] (Table 5). However, the Iino score has not been widely used in clinical practice due to the lack of validation and the difficulty in interviewing all patients before endoscopy due to syncope and antiplatelet medication.

We evaluated the usefulness of a scoring system for the risk of gastrointestinal bleeding using the GBS, AIMS65, MAP score, Modified GBS, and Iino score. The usefulness of the scoring system for gastrointestinal bleeding risk was evaluated. We extracted factors related to endoscopic treatment and created a new score that is simpler and more useful (Table 6). In this study, we evaluate the usefulness of the new score and the existing prediction score in predicting hemostatic therapy in NVUGIB, the accuracy of predicting mortality, and advocate the use of the new score.

Patient management

From January 2015 to December 2019, patients with symptoms of upper gastrointestinal bleeding, such as hematemesis, black stool, fainting, and anemia, who were transported to the National Disaster Medical Center of the National Hospital Organization or visited the outpatient clinic, and underwent upper gastrointestinal endoscopy within 6 hours of the visit, with upper gastrointestinal bleeding suspected based on symptoms and blood tests, were retrospectively reviewed. A total of 675 patients with suspected non-variceal upper gastrointestinal bleeding were included in the study, excluding patients with esophageal or gastric varices. Blood samples were taken immediately after the visit, and clinical symptoms were interviewed at the time of the visit or before treatment. The GBS, AIMS65, MAP score, Modified GBS, and Iino score were calculated to determine whether hemostatic treatment or other hemostatic treatment was necessary at the subsequent emergency endoscopy and whether death occurred afterwards. Endoscopic hemostasis was performed according to the guidelines of the Japanese Society of Gastrointestinal Endoscopy [11]. The hemostatic treatment of ulcerative lesions was performed according to Forrest's modified classification of active bleeding (Ia: eruptive bleeding, Ib: exudative bleeding) and IIa (with guttural bleeding vessels) among recent bleeding [16].

This retrospective study was approved by the Hospital Ethical Review Committee at the National Disaster Medical Center in December 2020 (National Disaster Medical Center: number 2020-17) and by the Clinical Research Ethics Review Committee at Nihon University Medical College Itabashi Hospital in February 2021 (Nihon University Medical College Itabashi Hospital: number RK-210209-9). The treatments performed were part of the current standard of care, and patient data were anonymized. Informed consent was exempted by Clinical Research Ethics Review Committee at Nihon University Medical College Itabashi Hospital due to the retrospective nature of the study. All methods were performed in accordance with the relevant guidelines and regulations.

Data collection

Patient data were collected by the authors and a full-time research clerk. The collected data included patient characteristics, clinical symptoms, blood tests, and treatment actions necessary to calculate each score. Endoscopic treatment, Interventional Radiology, surgery, and death were recorded (Table 1).

Statistical analysis

Optimal score thresholds for predicting low-risk patients who may not require hemostatic treatment were identified based on a sensitivity of >95%. The relationship between each item and endoscopic treatment was evaluated using binomial logistic analysis to determine which of the items was associated with hemostatic treatment. Significantly related factors were extracted, and a new score associated with hemostasis (HB score) was created by considering partial regression coefficients. Next, we stratified the number of patients who received hemostatic treatment or not according to the five existing gastrointestinal bleeding risk score systems (GBS, AIMS65, MAP score, Modified GBS, and Iino score) and the new score (HB score) according to each risk score system.

We also used area under the receiver operating characteristic curves (AUROC) to compare the ability of each of the existing scores to predict the need for endoscopic treatment with the new score. We validated the cutoff value of the new score and compared the ability of each existing score and the new score to predict subsequent death using AUROC.

Patient characteristics and baseline scores

A total of 675 patients were included in this study. The mean age was 72.3 years, and the male to female ratio was about 6:4. 24% had a history of cardiac disease, and 6% had a history of liver disease. 57% of the patients showed hemorrhage and black stools, 25% showed hematemesis, and 7% showed disturbance of consciousness. All 223 patients who underwent endoscopy required hemostatic treatment.2 underwent Interventional Radiology in addition to endoscopic treatment and 3 underwent surgical treatment. 19 patients died between the time of consultation and the time of hospitalization (Table 7).

Factors associated with gastrointestinal bleeding

The GBS consists of BUN, Hb, systolic blood pressure, pulse rate, hypovolemia/black stools, syncope, cardiac and cardiovascular disease [7]. AIMS65 consists of albumin, PT-INR, mental and consciousness disorders, systolic blood pressure, and age [8]. The Iino score consists of items such as systolic blood pressure, syncope, hematemesis, hemoglobin, BUN, eGFR, and antiplatelet agents [10]. We conducted a logistic analysis to evaluate whether the individual factors comprising the above three risk scoring systems were related to endoscopic hemostatic therapy in patients with gastrointestinal bleeding. Factors associated with hemostatic treatment were blood pressure (P = 0.0283), BUN (P <0.001), Hb (P = 0.0037), hematemesis (P = 0.0030), and pulse (P = 0.0137) were the factors directly related to hemostatic treatment in emergency endoscopy (9) (Table 8). Since there is no existing score that includes all of them, we created a new score model (Table 6).

Comparison of the existing score and the new score in terms of their ability to predict the need for hemostatic treatment and subsequent death during emergency endoscopy

Figure 1 represents the relationship between the number of patients undergoing hemostatic treatment and each of the 5 score analysed (Figure 1A-E). The ROC curve was used to evaluate the usefulness of the existing score and the new HB score (Table 9). In the existing model, GBS and Modified GBS showed higher discriminatory ability (AUROC: 0.721 and 0.7128, respectively; P < 0.001) in predicting hemostatic treatment in emergency endoscopy compared with other existing scores, and the new HB score was even more discriminatory (AUROC: 0.73). The optimal cutoff value for the HB score was 3 points (Figure 2). The HB score also showed higher prediction accuracy for subsequent deaths (AUROC: 0.6857. P <0.001), although it was inferior to AIMS65 (AUROC: 0.7070; P <0.001); P < 0.001) (Figure 3).

Hemostasis during emergency endoscopy has become easier than before with the advancement of instruments, and a high success rate has been maintained [17, 18]. However, spontaneous hemostasis may occur during emergency endoscopy, and endoscopic hemostasis is not necessary in all cases [19].

In cases of suspected acute gastrointestinal bleeding, early risk stratification to identify high-risk patients is important and is associated with the timing and need for subsequent endoscopy. In an international multicenter prospective study of more than 3,000 patients, the GBS was the best predictor of the need for hospital-based intervention or death. In an international multicenter prospective study of more than 3,000 patients, GBS was the best predictor of the need for hospital-based intervention or death [8]. A GBS score of 7 or higher indicated the need for endoscopic intervention. Another report showed that a GBS of 12 points or higher had a 90% specificity for predicting in-hospital mortality, and that a delay in endoscopy significantly increased the number of deaths in patients with a GBS of 12 points or higher [20]. GBS was useful for endoscopic intervention [21] and hemostatic treatment [22] among conventional scores. The GBS score correlates with therapeutic intervention and lethality as the score increases, but the actual score at which therapeutic intervention should be performed has not been determined. A systematic review of 16 reports confirmed that the GBS was better than other cutoff points and risk scores at identifying low-risk patients, but had very low specificity [10]. Therefore, it is difficult to accurately assess the risk of patients with upper gastrointestinal bleeding to determine what cases should be promptly endoscoped and treated [23].

Therefore, we conducted this study to examine the association between each existing score and the need for endoscopic or other hemostatic treatment. The GBS was found to be the most useful in predicting hemostatic treatment, as previously reported. However, there are many items in the GBS, and it is difficult to assess accurately in emergency medicine because it includes subjective factors such as fainting and the presence of cardiac and hepatic diseases. In addition, it is difficult to define cardiac and hepatic diseases, and it is a difficult question whether to include even minor diseases. In addition, it is difficult to define cardiac and hepatic diseases.

From our study, the factors associated with endoscopic treatment were hematemesis, heart rate (>100 beats/min), Hb (Hemoglobin: 10.0 g/dl or less), blood pressure (Blood pressure: 100 mmHg or less), BUN (Blood urea BUN (Blood urea nitrogen: 22.4 mg/dl or higher). Hematemesis can be observed from the surroundings and can be easily and objectively assessed by examining the oral cavity and perioral area. Therefore, the above five factors are all objective indicators. The measurement of each clinical index is simple and can be performed in actual clinical situations requiring emergency. Based on the above predictors and the partial regression coefficients, we proposed a new score with a total of 6 points, 2 points for BUN and 1 point for each of the other factors, and named this score the HB score, using the initial letters of each factor. The HB score can be used as a useful index compared to GBS and Modified GBS. The HB score has a cutoff value of 3 points, which is the maximum value based on the Youden index, as a guideline for hemostatic treatment. However, considering the clinical aspects, a score of 2 or higher, which has a certain degree of sensitivity, should be considered as an indication for urgent endoscopy and should be a subject for further study. Although the HB score is inferior to the AIMS65, it is also an excellent predictor of subsequent death.

There are, of course, some limitations to this study. First, it was an analysis of cases from a single institution. Also, patients with esophageal varices were not analyzed, so our conclusions cannot be applied to all patients with suspected upper gastrointestinal bleeding. However, there are few data on the use of scoring systems in patients with variceal bleeding, and the predictive power is low. In this study, we did not uniformly administer PPIs between the time of our consultation and the endoscopic intervention, although patients were taking PPIs regularly. North American and European guidelines suggest the use of high-dose PPIs before endoscopy as basic therapy to reduce the incidence of peptic ulcer [24-26]. Patients receiving PPIs prior to endoscopic intervention were significantly less likely to develop adverse outcomes and had significantly lower rates of rebleeding, upper gastrointestinal surgery, mortality, and length of hospital stay compared with patients who did not receive PPIs [27-29]. Potassium-competitive acid blockers (P-CABs) were approved in Japan in 2015, the first in the world. P-CABs do not require acid activation, are stable in acidic environments, and accumulate in secretory tubules in high density. This may be an issue for further study.

In this study, the primary endpoint was the presence or absence of hemostatic treatment in patients who underwent emergency endoscopy within 6 hours of presentation. A prospective observational study in Korea evaluated mortality and rebleeding rate 28 days after hospitalization in adult patients with GBS 7 or higher NVUGIB who underwent endoscopy within 6 hours of emergency department visit and those who underwent endoscopy within 6-48 hours. The mortality rate was significantly lower in the group that underwent endoscopy within 6 hours, but there was no difference in rebleeding between the two groups [30]. Another report examined whether emergency upper gastrointestinal endoscopy within 6 hours improved all-cause mortality at 30 days compared with early endoscopy within 6 to 24 hours in patients with stable upper gastrointestinal hemorrhage without persistent hemorrhage (GBS 12 points or higher). However, it was reported that there was no significant difference in all-cause mortality between the two groups [19]. There is controversy about when endoscopy should be performed in patients with suspected acute upper gastrointestinal bleeding [31]. Even if the new score can predict hemostatic therapy such as endoscopic treatment and mortality, it remains to be determined when therapeutic intervention can improve important clinical outcomes such as death.

Shung et al. validated a machine-learning model for hemostatic treatment requiring hospitalization and 30-day mortality in patients with upper gastrointestinal bleeding using a 24-item index that included demographics, comorbidities, medications, clinical characteristics, and blood sampling results. It showed higher AUC, sensitivity, and specificity than the conventional GBS and AIMS65 scores [32]. However, it requires many items and a detailed interview, and patients themselves often do not remember their medications and comorbidities accurately, so it is not suitable for urgent care at present. However, with the development of AI and digitalization, for more efficient collection of medical information, the Shung et al. score with higher sensitivity and specificity using more indicators may become necessary in the future.

In summary, we proposed a new score, the HB score, consisting of simple and objective indices in cases of suspected upper gastrointestinal bleeding. The HB score is useful in identifying high-risk patients with suspected upper gastrointestinal bleeding who require urgent hemostatic treatment including emergency endoscopy.

ASA: American Society of Anesthesiologists

AUROC: area under the receiver operating characteristic curves

BUN: blood urea nitrogen

GBS: Glasgow-Blatchford Score

HB: hemoglobin

HB score: score associated with hemostasis

NSAIDs: nonsteroidal anti-inflammatory drugs

NVUGIB: non-variceal upper gastrointestinal bleeding

P-CABs: Potassium-competitive acid blockers

PPI: Proton Pump Inhibitor

PT-INR: Prothrombin time-international normalized ratio

Ethics approval and consent to participate: This retrospective study was approved by the Hospital Ethical Review Committee at the National Disaster Medical Center in December 2020 (National Disaster Medical Center: number 2020-17) and by the Clinical Research Ethics Review Committee at Nihon University Medical College Itabashi Hospital in February 2021 (Nihon University Medical College Itabashi Hospital: number RK-210209-9). The treatments performed were part of the current standard of care, and patient data were anonymized. Informed consent was exempted by Clinical Research Ethics Review Committee at Nihon University Medical College Itabashi Hospital due to the retrospective nature of the study. All methods were performed in accordance with the relevant guidelines and regulations.

Consent for publication: Not applicable.

Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests: The authors declare that they have no competing interests.

Funding: This work received no funding.

Authors' contributions: YS- Writing - Original draft; TA- Data Curation; NK, TK, IS, SO- Conceptualization, Methodology; KO- Validation; MM- Supervision, Project administration. All authors read and approved the final manuscript.

Acknowledgements: Editorial support, in the form of medical writing, assembling tables and creating high-resolution images based on authors’ detailed directions, collating author comments, copyediting, fact checking, and referencing, was provided by Editage, Cactus Communications.

Lanas A, García-Rodríguez LA, Polo-Tomás M, Ponce M, Alonso-Abreu I, Perez-Aisa MA, et al. Time trends and impact of upper and lower gastrointestinal bleeding and perforation in clinical practice. Am J Gastroenterol. 2009;104:1633-41.
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;107:1190-6.
Oakland K. Changing epidemiology and etiology of upper and lower gastrointestinal bleeding. Best Pract Res Clin Gastroenterol. 2019;42–43:101610.
Sung JJ, Chiu PW, Chan FKL, Lau JY, Goh KL, Ho LH, et al. Asia-Pacific working group consensus on non-variceal upper gastrointestinal bleeding: an update 2018. Gut. 2018;67:1757-68.
Gralnek IM, Stanley AJ, Morris AJ, Camus M, Lau J, Lanas A, et al. Endoscopic diagnosis and management of nonvariceal upper gastrointestinal hemorrhage (NVUGIH): European Society of Gastrointestinal Endoscopy (ESGE) Guideline - Update 2021. Endoscopy. 2021;53:300-32
Blatchford O, Murray WR, Blatchford M. A risk score to predict need for treatment for upper-gastrointestinal haemorrhage. Lancet. 2000;356:1318-21.
Saltzman JR, Tabak YP, Hyett BH, Sun X, Travis AC, Johannes RS. A simple risk score accurately predicts in-hospital mortality, length of stay, and cost in acute upper GI bleeding. Gastrointest Endosc. 2011;74:1215-24.
Stanley AJ, Laine L, Dalton HR, Ngu JH, Schultz M, Abazi R, et al. Comparison of risk scoring systems for patients presenting with upper gastrointestinal bleeding: international multicentre prospective study. BMJ. 2017;356:i6432.
Robertson M, Majumdar A, Boyapati R, Chung W, Worland T, Terbah R, et al. Risk stratification in acute upper GI bleeding: comparison of the AIMS65 score with the Glasgow-Blatchford and Rockall scoring systems. Gastrointest Endosc. 2016;83:1151-60.
Ramaekers R, Mukarram M, Smith CA, Thiruganasambandamoorthy V. The predictive value of Preendoscopic Risk Scores to predict adverse outcomes in emergency department patients with upper gastrointestinal bleeding: A systematic review. Acad Emerg Med. 2016;23:1218-27.
Fujishiro M, Iguchi M, Kakushima N, Kato M, Sakata Y, Hoteya S, et al. Guidelines for endoscopic management of non-variceal upper gastrointestinal bleeding. Dig Endosc. 2016;28:363-78.
Redondo-Cerezo E, Vadillo-Calles F, Stanley AJ, Laursen S, Laine L, Dalton HR, et al. MAP(ASH): A new scoring system for the prediction of intervention and mortality in upper gastrointestinal bleeding. J Gastroenterol Hepatol. 2020;35:82-9.
Cheng DW, Lu YW, Teller T, Sekhon HK, Wu BU. A modified Glasgow Blatchford Score improves risk stratification in upper gastrointestinal bleed: a prospective comparison of scoring systems. Aliment Pharmacol Ther. 2012;36:782-9.
Quach DT, Dao NH, Dinh MC, Nguyen CH, Ho LX, Nguyen ND, et al. The performance of a Modified Glasgow Blatchford Score in predicting clinical interventions in patients with acute nonvariceal upper gastrointestinal bleeding: A Vietnamese prospective multicenter cohort study. Gut Liver. 2016;10:375-81.
Iino C, Mikami T, Igarashi T, Aihara T, Ishii K, Sakamoto J, et al. Evaluation of scoring models for identifying the need for therapeutic intervention of upper gastrointestinal bleeding: A new prediction score model for Japanese patients. Dig Endosc. 2016;28:714-21.
Heldwein W, Schreiner J, Pedrazzoli J, Lehnert P. Is the Forrest classification a useful tool for planning endoscopic therapy of bleeding peptic ulcers? Endoscopy. 1989;21:258-62.
Hagel AF, Albrecht H, Nägel A, Vitali F, Vetter M, Dauth C, et al. The application of hemospray in gastrointestinal bleeding during emergency endoscopy. Gastroenterol Res Pract. 2017;2017:3083481.
Jacques J, Legros R, Chaussade S, Sautereau D. Endoscopic haemostasis: an overview of procedures and clinical scenarios. Dig Liver Dis. 2014;46:766-76.
Lau JYW, Yu Y, Tang RSY, Chan HCH, Yip HC, Chan SM, et al. Timing of endoscopy for acute upper gastrointestinal bleeding. N Engl J Med. 2020;382:1299-1308.
Mujtaba S, Chawla S, Massaad JF. Diagnosis and management of non-variceal gastrointestinal hemorrhage: A review of current guidelines and future perspectives. J Clin Med. 2020;9:402.
Martínez-Cara JG, Jiménez-Rosales R, Úbeda-Muñoz M, de Hierro ML, de Teresa J, Redondo-Cerezo E. Comparison of AIMS65, Glasgow-Blatchford score, and Rockall score in a European series of patients with upper gastrointestinal bleeding: performance when predicting in-hospital and delayed mortality. United European Gastroenterol J. 2016;4:371-9.
Laursen SB, Hansen JM, Schaffalitzky de Muckadell OB. The Glasgow Blatchford score is the most accurate assessment of patients with upper gastrointestinal hemorrhage. Clin Gastroenterol Hepatol. 2012;10:1130-1135.e1.
Sengupta N. Challenges to risk determination for patients with upper gastrointestinal bleeding. Clin Gastroenterol Hepatol. 2019;17:1037-9.
Barkun AN, Bardou M, Kuipers EJ, Sung J, Hunt RH, Martel M, et al. International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding. Ann Intern Med. 2010;152:101-13.
Laine L, Jensen DM. Management of patients with ulcer bleeding. Am J Gastroenterol. 2012;107:345-61.
Gralnek IM, Dumonceau JM, Kuipers EJ, Lanas A, Sanders DS, Kurien M, et al. Diagnosis and management of nonvariceal upper gastrointestinal hemorrhage: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy. 2015;47:a1-46.
Keyvani L, Murthy S, Leeson S, Targownik LE. Pre-endoscopic proton pump inhibitor therapy reduces recurrent adverse gastrointestinal outcomes in patients with acute non-variceal upper gastrointestinal bleeding. Aliment Pharmacol Ther. 2006;24:1247-55.
Marmo R, Soncini M, Bucci C, Zullo A; Gruppo Italiano per lo Studio dell'Emorragia Digestiva (GISED). Pre-endoscopic intravenous proton pump inhibitors therapy for upper gastrointestinal bleeding: A prospective, multicentre study. Dig Liver Dis. 2021;53:102-6.
Laursen SB, Leontiadis GI, Stanley AJ, Møller MH, Hansen JM, Schaffalitzky de Muckadell OB. Relationship between timing of endoscopy and mortality in patients with peptic ulcer bleeding: a nationwide cohort study. Gastrointest Endosc. 2017;85:936-44.e3.
Cho SH, Lee YS, Kim YJ, Sohn CH, Ahn S, Seo DW, et al. Outcomes and role of urgent endoscopy in high-risk patients with acute nonvariceal gastrointestinal bleeding. Clin Gastroenterol Hepatol. 2018;16:370-7.
Jung K, Moon W. Role of endoscopy in acute gastrointestinal bleeding in real clinical practice: An evidence-based review. World J Gastrointest Endosc. 2019;11:68-83.
Shung DL, Au B, Taylor RA, Tay JK, Laursen SB, Stanley AJ, et al. Validation of a machine learning model that outperforms clinical risk scoring systems for upper gastrointestinal bleeding. Gastroenterology. 2020;158:160-7.

Table 1. Glasgow - Blatchford score

Item	Standard	Score
BUN, (mmol / L) (Value in mg / dl)	＜6.5　　　(＜18.2）	0
	6.5-7.9　(18.2-22.4）	2
	8.0-9.9　(22.5-28.0）	3
	10.0-24.9(28.1-69.0）	4
	≧25.0　　(≧70)	6
Hemoglobin, male (g / dl)	12-12.9	1
	10-11.9	3
	<10	6
Hemoglobin, female (g / dl)	10-11.9	1
Hemoglobin, female (g / dl)	<10	6
Systolic blood pressure (mmHg)	100-109	1
	90-99	2
	<90	3
Other indicators	Pulse ≧ 100	1
	Bloody stool	1
	Fainting	2
	Liver disease	2
	Heart disease	2

The need for treatment for upper gastrointestinal bleeding is stratified using BUN, Hb, blood pressure, and other indicators, and evaluated on a scale of 0-23.

Table 2. AIMS65

Item	Standard	Score
Albumin (mg / dl)	< 3.0	1
PT-INR	＞1.5	1
Disturbance of consciousness		1
Systolic blood pressure (mmHg)	＜90	1
Age (years)	＞65	1

In order to easily predict the prognosis of upper gastrointestinal bleeding, 5 indicators of albumin, PT-INR, impaired consciousness, blood pressure, and age are used to evaluate on a score of 0-5.

Table 3. MAP score

Item	Standard	Score
Disturbance of consciousness	Glasgow　Coma Scale < 15	１
American Society of Anesthesiologists (ASA) score	>2	１
Pulse (times / minute)	> 100	１
Albumin (mg / dl)	＜2.5	２
Systolic blood pressure (mmHg)	<90	２
Hemoglobin (g / dl)	＜10	２

In order to predict treatment intervention including endoscopy and mortality in upper gastrointestinal bleeding, evaluation is performed on a score of 0-9 using the above items.

Table 4. Modified Glasgow - Blatchford score

Item	Standard	Score
BUN, (mmol / L) (Value in mg / dl)	＜6.5　　　(＜18.2）	0
	6.5-7.9　(18.2-22.4）	2
	8.0-9.9　(22.5-28.0）	3
	10.0-24.9(28.1-69.0）	4
	≧25.0　　(≧70)	6
Hemoglobin, male (g / dl)	12-12.9	1
	10-11.9	3
	<10	6
Hemoglobin, female (g / dl)	10-11.9	1
Hemoglobin, female (g / dl)	<10	6
Systolic blood pressure (mmHg)	100-109	1
	90-99	2
	<90	3
Pulse (times / minute)	≧100	1

Glasgow --Blatchford score 0-16 using only 5 indicators of BUN, Hb, systolic blood pressure, and pulse.

Table 5. Iino score

Item	Standard	Score
Systolic blood pressure (mmHg)	<100	２
Syncope		２
Hematemesis		３
Hemoglobin (g / dl)	＜10	１
BUN (mmol / L)	≧22.4	２
eGFR (mL / min / 1.73m2)	≧60	-２
Oral antiplatelet drug		-２

In order to predict the indication of endoscopic treatment for upper gastrointestinal bleeding, the above items are used to evaluate on a score of -4 to 10.

Table 6. New score (score associated with hemostasis)

Item	Standard	Score
Hematemesis	<100	１
Heart rate(times / minute)	≧100	１
Blood pressure (blood pressure, mmHg or less)	≦100	１
Hb (g / dl or less)	≦10	１
BUN (mg / dl or mor	≧22.4	２

In order to predict the indication of hemostasis treatment including endoscopic treatment in upper gastrointestinal bleeding, the above items are used to evaluate on a score of 0-6.

Table 7. Baseline data of patients undergoing emergency endoscopy in 2015-2019

Item	Numerical value
Mean age (years)	72.3 ±14.5
Men/women	414(61%)/261(39%)
Systolic/diastolic BP	123.3±23.7/68.2±13.4
Pulse	84.1±18.8
History (heart/liver disease)	164(24%)/43(6%)
Antiplatelets/anticoagulants	133(20%)/124(18%）
Hb	9.1±2.7
BUN	36.4±28.5
Alb	2.8±1.4
eGFR	59.7±31.7
INR	1.0±0.8
Impaired consciousness	47(7.0%)
Melena	383(56.7%)
Fainting	28(4.1%)
Hematemesis	167(24.7%)
Helicobacter pylori positive	107(15.9%)
NSAIDs takers	61(9.0%)
Steroid users	16(2.4%)
PPI oral	187(27.7%)
Helicobacter pylori eradication treatment history	6(0.9%)
Endoscopic hemostasis therapist (including unsuccessful)	223(33.0%)
Interventional Radiology	2(0.3%)
Surgical treatment	3(0.4%)
death	19(2.8%)

Table 8. Binary logistic regression model to determine the independent association of factors involved in gastrointestinal bleeding on the undertaking hemostatic treatment upon endoscopy

				Partial regression coefficient 95% confidence interval			Odds ratio 95% confidence interval		Partial regression coefficient significance test
Variable	Partial regression coefficient	Standard error	Standardized partial regression coefficient	Lower endpoint	Upper endpoint	Odds ratio	Lower endpoint	Upper endpoint	Wald	Degree of freedom	P-value	：P<0.05 *：P<0.01
Liver disease	-0.3078	0.3954	-0.0752	-1.0828	0.4673	0.7351	0.3386	1.5956	0.6058	1	0.4364
Heart failure	-0.1034	0.2442	-0.0444	-0.5821	0.3752	0.9017	0.5587	1.4553	0.1794	1	0.6719
Antiplatelets	0.0382	0.2322	0.0152	-0.4170	0.4934	1.0389	0.6590	1.6379	0.0270	1	0.8694
Anticoagulants	0.1142	0.2555	0.0442	-0.3866	0.6150	1.1209	0.6793	1.8496	0.1996	1	0.6550
Mental illness, impaired consciousness	0.1122	0.3422	0.0285	-0.5586	0.7829	1.1187	0.5720	2.1878	0.1074	1	0.7431
Melena	0.3794	0.1947	0.1880	-0.0022	0.7610	1.4614	0.9978	2.1403	3.7970	1	0.0513
Fainting	0.7034	0.4236	0.1402	-0.1268	1.5335	2.0205	0.8809	4.6346	2.7573	1	0.0968
Hematemesis	0.6682	0.2254	0.2883	0.2265	1.1098	1.9506	1.2542	3.0339	8.7908	1	0.0030	**
Pulse	0.5729	0.2324	0.2124	0.1173	1.0285	1.7734	1.1245	2.7968	6.0746	1	0.0137	*
Albumin (≤3.0)	-0.0054	0.1892	-0.0026	-0.3762	0.3654	0.9946	0.6865	1.4411	0.0008	1	0.9773
INR (≥1.5)	-0.0233	0.1948	-0.0110	-0.4051	0.3584	0.9769	0.6669	1.4310	0.0144	1	0.9046
Blood pressure (≤90)	0.6954	0.4924	0.1478	-0.2698	1.6605	2.0044	0.7635	5.2620	1.9941	1	0.1579
Age (≥65 years)	0.0089	0.2279	0.0039	-0.4377	0.4555	1.0090	0.6455	1.5770	0.0015	1	0.9688
Blood pressure (≤100)	0.6599	0.3008	0.2233	0.0703	1.2495	1.9345	1.0728	3.4885	4.8116	1	0.0283	*
Hb (≤10.0）	0.5902	0.2035	0.2848	0.1914	0.9890	1.8043	1.2110	2.6885	8.4148	1	0.0037	**
BUN (≥22.4）	1.0484	0.2057	0.5133	0.6453	1.4515	2.8530	1.9065	4.2694	25.9843	1	P < 0.001	**
eGFR (≥60）	0.0091	0.2003	0.0045	-0.3835	0.4017	1.0091	0.6815	1.4944	0.0021	1	0.9637

Table 9. ROC curve data (score associated with hemostasis)

Reference point	TPF (True Positive Fraction）	FPF (false positive Fraction）	Odds ratio	Specificity
0	1.0000	1.0000	-	0
1	0.9821	0.8982	6.2032	0.1018
2	0.8879	0.6062	5.1451	0.3938
3	0.7534	0.4071	4.4490	0.5929
4	0.3812	0.1018	5.4364	0.8982
5	0.0807	0.0111	7.8498	0.9889
6	0.0090	0.0022	4.0814	0.9978

The optimal cutoff value for the HB score was 3 points

No competing interests reported.

PDF

Editorial decision: Major revision
06 Jun, 2022
Reviews received at journal
04 Jun, 2022
Reviewers agreed at journal
01 Jun, 2022
Reviewers agreed at journal
31 May, 2022
Reviewers invited by journal
31 May, 2022
Editor assigned by journal
31 May, 2022
Editor invited by journal
31 May, 2022
Submission checks completed at journal
31 May, 2022
First submitted to journal
18 May, 2022

You are reading this latest preprint version

Prediction of the need for emergency endoscopic treatment for upper gastrointestinal bleeding and new score model: A retrospective study

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Discussion

Conclusions

List Of Abbreviations

Declarations

References

Tables

Competing Interests

Status:

Version 1