Endoscopic findings and predictors of outcome among patients with non-variceal upper gastrointestinal bleeding attending Benjamin Mkapa Hospital in Dodoma, Tanzania

doi:10.21203/rs.3.rs-4963591/v1

Download PDF

Research Article

Endoscopic findings and predictors of outcome among patients with non-variceal upper gastrointestinal bleeding attending Benjamin Mkapa Hospital in Dodoma, Tanzania

https://doi.org/10.21203/rs.3.rs-4963591/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Non-variceal upper gastrointestinal bleeding (NVUGIB) is a medical emergency and is associated with poor outcomes. In our setting, there are increased trends of upper gastrointestinal diseases including liver cirrhosis with challenges in management; however, the published data are limited. This study aimed to assess the endoscopic findings and predictors of outcomes of patients with NVUGIB attending Benjamin Mkapa Hospital in Dodoma, Tanzania.

Methods

This prospective longitudinal study was conducted between October 2023 and April 2024. All eligible patients with NVUGIB attending Benjamin Mkapa Hospital were recruited. Consented participants who underwent endoscopic procedures aged 18 years or older were included. Baseline sociodemographic and clinical characteristics and endoscopic procedures were performed during inception. Patients with NVUGIB were followed up for 60 days to assess the outcomes. Data were summarized by descriptive analysis then, presented as a median and interquartile range (IQR) for continuous variables, and categorical variables were analyzed by Fisher exact and Mann-Whitney U tests and reported as numbers and proportions. Binary regression analysis used to assess the predictors of rebleeding within 60 days and reported using odds ratio. A two-tailed p-value of less than 0.05 was considered statistically significant.

Results

Among 124 patients, approximately 57.3% were male with a median age of 44 (IQR 27–61) years. Approximately, 70.2% had H. pylori infections and 42.7% were alcoholic. Median Haemoglobin level was 10.6 (IQR 6.1–15.9); median platelet count accounting for 67 (IQR 43–480) while median SBP was 94 (IQR 79–141 mmHg and median DBP of 64 (48–82) mmHg. Prevalence of NVUGIB was 24.2%. Among NVUGIB patients, 42% had Gastritis and 1.8% had esophageal tumors. Rebleeding was prevalent in 21.8%, whereas, 2.4% required an ICU admission and death was observed in 4%. Predictors of rebleeding included low Haemoglobin (AOR 2.24; 95% CI, 1.39–2.98, p = 0.002); Forest class 1 (AOR 2.51; 95% CI, 1.34–4.02, p = 0.031); Forest class 2 (AOR 1.81; 95% CI, 1.18–2.82, p = 0.044) and thrombocytopenia (AOR 1.43; 95% CI, 1.15–2.71, p = 0.011).

Conclusion

Rebleeding is a common outcome among patients with NVUGIB. Therefore, intensive management is crucial to improve patient outcomes. However, future research is warranted.

Endoscopic findings

predictors

outcome

non-variceal upper gastrointestinal bleeding

Upper gastrointestinal bleeding is characterized by bleeding between the esophagus and the ligament of Treitz, in any location [1]. It requires immediate medical attention and hospitalization due to its life-threatening nature, accounting for a significant number of global fatalities [2].UGIB is classified into variceal and non-variceal upper gastrointestinal bleeding. Non-variceal bleeding refers to bleeding that develops in the esophagus, stomach, or proximal duodenum [3].The most common cause of upper gastrointestinal bleeding is peptic ulcer which is responsible for about 31–67% of all cases, followed by erosive disease and Variceal bleeding [4].

Globally there is a scarcity of data on endoscopic findings of NVUGB, however several studies done in different countries different patterns have been observed. In two studies done in Saudi Arabia, the endoscopic findings in NVUGIB patients included esophagitis, gastric erosions, duodenal ulcers, gastric ulcers, and less commonly, tumors [5] and another study which highlighted the commonest pattern being gastritis 52.6%, gastric ulcer 17.9%, esophagitis 15.6%, gastric and duodenal ulcers 12% [6, 7]. A study from Brazil reported a significant frequency of NVUGIB secondary to peptic ulcer disease, while tumors were a less common cause [8]. Other studies have identified esophagitis, gastritis, peptic ulcer disease, and normal endoscopic pattern in NVUGIB patients [6, 9, 10].

In Africa, the following pattern was observed in Nigeria and Uganda whereby the commonest pattern was gastritis, peptic ulcer, erosive mucosa, malignancies, and esophagitis. [11, 12].In a study done in Tanzania, esophageal varices (70%) were the common pattern while in non-Variceal the commonest pattern was peptic ulcer [13–15].

It has been established that the outcomes of non-variceal bleeding are mortality and rebleeding. According to the studies done in the USA, by [16, 17] the mortality was 2% − 16%, however, re-bleeding was assessed in one study and was found to be 18.6%. Other studies done in Korea, Portugal, Turkey, and Pakistani found the re-bleeding ranges from 6.9–21.7% and mortality ranges from 4.2–16% [18, 19].In Africa there is a paucity of data, however, according to a study done in Zimbabwe the mortality was 13% among both variceal and non-variceal bleeding, and in studies done in Tanzania, the mortality was ranged between 17–33.5%. [15, 20]

Re-bleeding is one of the significant predictors of death and influences other important endpoints like surgery, massive blood transfusion, and prolonged hospital stay (Shaheen 2023). Studies have highlighted the following predictors of re-bleeding to be clinical symptoms like hematemesis, low hemoglobin level (< 8g/dl), ulcer size, inexperienced endoscopist, location of ulcer, Forest class, Rockall score of > 4, age > 40 years, need of > 15 cc of epinephrine, comorbidity with liver disease and chronic kidney disease, NSAIDs, altered mental status [2, 21] .

Non-variceal upper gastrointestinal bleeding is an emergency that has considerable morbidity and mortality especially when it is acute [22]. Among the established outcomes of NVUGB are prolonged hospital stay, mortality, and re-bleeding. Re-bleeding is one of the most important predictors of morbidity and mortality. Identifying predictors that are likely to increase the risk of re-bleeding is critical in the management of the patients with NVUGB to prevent mortality and morbidity due to re-bleeding, however little is documented in our setting with regard to nonvariceal bleeding, outcome, and predictors of re-bleeding. This study was undertaken to assess the endoscopic findings, outcomes, and predictors of re-bleeding inpatients attending at Dodoma referral hospitals in Tanzania.

Study Design

This study employed a prospective longitudinal study to assess the endoscopic findings and predictors of nonvariceal upper gastrointestinal bleeding among patients attending Benjamin Mkapa Hospital (BMH) in Dodoma, Tanzania from October 2023 to April 2024.

Study setting

The study was conducted at BMH in the capital city of Dodoma, Tanzania. The study area is conveniently selected because is one of the tertiary public hospitals that offers endoscopic services to the majority of individuals whereas normal individuals can afford the services even without having insurance. This Hospital serves the population of the Dodoma region and nearby districts from neighbouring regions. The places include; Dodoma urban, Bahi, Chemba, Kongwa, Kondoa, and Mpwapwa districts, and some of the nearby villages and districts of Singida, Iringa, and Morogoro regions. Also, BMH functions as an academic, research, and teaching hospital affiliated with the University of Dodoma. Our facility (at the medical clinic) screened an average of 137 patients for UGIB per month. BMH serves an average number of 60 to 90 UGIB patients per year (according to BMH monthly routine report).

Study Population

This study population includes all patients with upper gastrointestinal bleeding within the last 14 days and confirmed with OGD, aged 18 years and above, attended at both outpatient and inpatient BMH in Dodoma, Tanzania.

Sample Size

The sample size was determined by using the expected proportion in the population, based on a previous study which is 10% [20]. Therefore, a minimum of 100 patients were required to meet the study objectives.

Inclusion Criteria

All patients aged 18 years and above with UGIB within the last 14 days and confirmed by the OGD.

Able to provide informed consent or a proxy consent from a close relative in case the patient is incapable.

Exclusion Criteria

All patients confirmed with both VUGIB & NVUGIB during endoscopy.

All patients with UGIB related to trauma.

All patients with incomplete medical information.

Sampling Technique

A convenient serial sampling technique was performed. A convenient serial sampling technique was used in the enrollments of adult patients with UGIB who then underwent endoscopic procedures for confirmation. The patients with UGIB who were available during the data collection exercise and who voluntarily willingly signed the written informed consent were eligible for this study. The process of enrollment was conducted with the help of the gastroenterology unit in charge by introducing the principal investigator and RAs (who collected data) to patients with UGIB. The enrollment was done by simple randomly between the inpatients and outpatients at the gastroenterology unit at BMH. During the data collection process, the patients were informed about the research's objectives and the significance of sharing their medical information. Following this introduction, the data collection started until the desired sample size was attained.

Data collection tools and methods

Data were collected from the study participants who met inclusion criteria through face-to-face interviews using a standard questionnaire. The data collection exercise was conducted by recruited research assistants. The questionnaire was prepared in English and translated into Kiswahili for easy participants to understand. Before the interview, participants were informed about the purpose of the study, and RAs asked for verbal consent to provide information. Those who voluntarily agreed were invited to participate in the study. The questionnaires included Socio-demographic characteristics, clinical characteristics, predictors of outcome, workouts, and management provided.

Study procedure

All patients with NVUGIB who met the inclusion criteria of the study were explained on the aim, safety, and benefits of the study and were requested to sign a written informed consent form. For those who were unable to read and write, researchers or research assistants read for them and they signed the form by using a thumb sign. History taking, general examination, and laboratory investigations taken after they consent to participate in this study. The endoscopic procedure was done and the patient was managed according to endoscopic findings together with other laboratory investigations Follow-up was done within 60 days for assessment of death, rebleeding, or ICU admission for inpatient and outpatient those with H. pylori-positive was started on triple therapy for 14 days and PPIs four weeks and free from medication for two weeks and then tested for cure.

Endoscopic Procedures

A white-light endoscope with great definition was used for routine upper endoscopy. The model 20017800 Storz Karlstorz from Germany was utilized. Patients are normally put on their left side with their neck extended forward. Topical pharyngeal anesthetic was performed using benzocaine spray or gel; however, if the patient did not tolerate it, intravenous sedation was employed. Before endoscopy, all patients or their relatives provided written informed consent that addressed potential risks, benefits, and treatment options during the procedure. To ensure optimal equipment orientation, a pre-procedure simulation of the up-and-down maneuver required to transfer the endoscope from the mouth to the upper esophageal sphincter is indicated.

The endoscope is passed posteriorly toward the upper esophageal sphincter, which is located 15 to 18 cm from the incisors and at the level of the thyroid cartilage. The upper esophageal sphincter is passed under direct view, frequently with the assistance of insufflation, a modest chin lift, and a small amount of pressure. After intubation, the tubular esophagus is examined. The esophagus is inspected both when the scope is inserted and after the tool is removed after inspecting the stomach and duodenum. To achieve full visualization, withdrawal was done carefully and with enough CO2 insufflation. Important criteria include the color of the mucosa and evidence of erythema, erosions, ulcers, strictures, rings, webs, varices, or diverticula. Biopsies were obtained for histology if appropriate. The endoscopic surgeon reported the finding promptly. (Rajni Ahlawat; et all 2023)

Laboratory procedure and data collection

Stool Antigen Testing (SAT) the non-invasive diagnostic method was used for establishing the presence of H. pylori it has a sensitivity of 95–100% and a specificity of 90%. The test was done first day of enrollment and at eight weeks from day one of treatment for assessing cure this was done for only positive the protocol was adhered

Full blood count

(FBC) This was done mainly to assess the hematological parameters with the interest of assessing hemoglobin level and platelets. Blood samples were collected by experienced nurses and doctors who were available at the post during the study period by following all sterile guidelines. The EDTA tube with the sample was labeled with the identification number of the patient and promptly sent to the laboratory. The machines. The calibrated XN 2000 SYMEX machine. From India. The result was then interpreted by the consulted hematologist and then recorded by the researcher

Study variable and definition:

Independent variables:

Social Demographic Characteristics

Age, Sex, residence, marital status, occupation/employment, and Education level.

Clinical characteristics:

Presenting Symptoms: (hematemesis, melena, hematochezia)
Comorbidities: (Hypertension, Diabetes mellitus, and other cardiovascular disease)
Medication use:(NSAIDs, Anticoagulants, Antiplatelets)
Risk Scores,
Pre-existing Medical Conditions,

Laboratory tests

Helicobacter pylori stool antigen, FBP (haemoglobin), AST, ALT, INR, serum creatinine, and BUN.

Intermediate variable

Source of Bleeding: (Peptic Ulcer Disease, Gastric erosion, duodenal ulcer)

Forrest Classification: (Forrest 1, 2, 3) Endoscopic Intervention, Endoscopic Therapy Use of Blood Transfusion, Surgical Intervention:

Dependent/Outcome variables: rebleeding, Death, ICU admission,

Predictors of rebleeding: assessed within 60 days

Data Processing and Analysis

Microsoft Excel was employed for data entry and data cleaning. IBM-SPSS program version 26 was used for data analysis under specific objectives. Baseline sociodemographic characteristics, clinical characteristics, and endoscopic procedures were performed during inception. Data were summarized by descriptive analysis and then, presented as a median and interquartile range (IQR) for continuous non-parametric variables, and categorical variables were analyzed by Fisher exact and Mann-Whitney U tests and reported as numbers and proportions. Outcomes were assessed on 2 weekly basis for 60 days and then were reported by descriptive analysis. Thereafter, binary regression analysis was used to assess the predictors of rebleeding assessed at 60 days and was reported by using odds ratio (OR). For univariate analysis, all variables fitted with a p-value of less than 0.2 were then exposed to multivariate analysis. A two-tailed p-value of less than 0.05 was considered statistically significant

In this study, a total of 512 patients were screened for UGIB after they presented with gastrointestinal symptoms. After excluding 388 patients for various reasons, approximately 124 patients had NVUGIB Fig. 1. Additionally, the prevalence of NVUGIB was found to be 24.2% (124/512). Moreover, 124 patients who met the inclusion criteria were followed up on 2 weekly bases via phone calls or during clinic visits for assessment of outcomes.

Baseline and clinical characteristics of the study population

In this study involving 124 patients, approximately 57.3% were male with a median age of 44 (IQR 27–61) years Table 1. At baseline, most of the patients with NVUGIB approximately 69.4% (86/124) presented with melena whereas, 81.5% (101/127) of the patients had upper abdominal pain. Furthermore, patients with H. pylori infections accounted for 70.2% (87/124) and 42.7% (53/124) were alcoholic. Median Haemoglobin level was 10.6 (IQR 6.1–15.9); median platelet count accounting for 67 (IQR 43–480) while a median SBP was 94 (IQR 79–141 mmHg and median DBP of 64 (48–82) mmHg.

Table 1: Outcomes of 124 patients with NVUGIB within 60 days

Variables	Numbers (%)
Life Status
Alive	119 (96)
Death	5 (4)
Rebleeding
No	97 (78.2)
Yes	27 (21.8)
ICU admission
No	121 (97.6)
Yes	3 (2.4)

Abbreviation: ICU-intensive care unit

Endoscopic findings of NVUGIB of the 124 study patients

Of 124 patients with NVUGIB, nearly half of the patients had Gastritis followed by duodenal ulcers accounting for 17.2% (21/124) and approximately 1.8% (2/124) had Esophageal tumours Fig. 2.

Treatment modality of NVUGIB of 124 study patients

As revealed in Fig. 3, the majority of patients 94 (71.2) of NVUGIB patients use PPI treatment, indicating the foundation of NVUGIB management. About 10 (7.6) of NVUGIB patients received a blood transfusion, indicating management of blood loss and anemia. Also, 87 (44.8) of NVUGIB patients tested with triple therapy for H. pylori infection that includes PPI and two antibiotics. A small percentage of 3 (2.3) of NVUGIB patients underwent surgical intervention, which was only reserved when endoscopic treatment failed or when uncontrollable bleeding occurred.

Outcomes of Patients with non-variceal upper gastrointestinal bleeding

Our findings revealed that rebleeding was present in 21.8% (27/124) whereas, approximately 2.4% (3/124) required an ICU admission and death was observed in 4% (5/124) Table 2. The small likelihood of death and ICU admission was due to modifiable effects explained by treatment modalities.

Table 2: Baseline sociodemographic and clinical characteristics of 124 study participants

Variables	Median (IQR) or Number (%)
Age in years	44 (IQR 27-61)
≤60	99 (79.8)
>60	25 (20.2)
Sex
Female	53 (42.7)
Male	71 (57.3)
Marital Status
Married	118 (95.2)
Single	6 (4.8)
Level of education
None	11 (8.9)
Primary	66 (53.2)
Secondary	42 (33.9)
University	5 (4.0)
Employment status
Employed	96 (77.4)
Unemployed	28 (22.6)
Hematemesis
No	89 (71.8)
Yes	35 (28.2)
Melena
No	38 (30.6)
Yes	86 (69.4)
Upper abdominal pain
No	23 (18.5)
Yes	101 (81.5)
Current alcohol consumption
No	71 (57.3)
Yes	53 (42.7)
Diabetes mellitus
No	114 (91.9)
Yes	10 (8.1)
Volume of alcohol/week in units
Female: ≤ 4	1 (4.2)
> 4	10 (41.7)
Male: ≤ 10	2 (8.3)
> 10	11 (45.8)
Duration of NSAIDs use (in months)
<3	83 (66.9)
≥3	41 (33.1)
PPI’s therapy
No	79 (63.7)
Yes	45 (36.3)
Cigarette smoking
No	105 (84.7)
Yes	19 (15.3)
Ischaemic heart disease
No	113 (91.1)
Yes	11 (8.9)
Renal diseases
No	119 (96)
Yes	5 (4)
Blood pressure measurement
DBP	64 (48-82)
SBP	94 (79-141)
Pulse Rate	93 (66-122)
Respiratory Rate	18 (12-25)
*H. Pylori Infection* (HPI)
Negative	95 (76.6)
Positive	29 (23.4)
White blood cells (10⁹/L)
High (≥11)	10 (8.1)
Normal (<11)	114 (91.9)
Haemoglobin level, g/dl
Low (<11)	116 (93.5)
Normal (≥11)	8 (6.5)
MCV
Low (<80)	63 (50.8)
Normal (80-100)	61(49.2)
Platelet counts (10⁹ /L)	67 (53-480)
Glasgow Blatchford Score
≤3	116 (93.5)
>3	8 (6.5)
Coagulation indices (PT/INR)
Normal	93 (75.2)
High (≥1.11)	31 (24.8)
Serum creatinine level, in mmol/L	113 (76-189)
Blood urea nitrogen, in mmol/L	6.4 (2.9-10)

Abbreviations: PPI-proton pump inhibitors; MCV-mean corpuscular volume; PT/INR-international normalized ratio; NSAIDS-non-steroidal anti-inflammatory drugs.

Table 3: Clinical profile of 27 study patients with rebleeding

Variables	No-rebleeding	Rebleeding	*p-value
Age, in years			0.017
≤60	46 (47.4)	17 (62.9)
>60	51 (52.5)	10 (37.1)
Cigarette smoking			0.603
No	83 (85.6)	22 (81.5)
Yes	14 (14.4)	5 (18.5)
Current alcohol consumption			0.189
No	78 (80.4)	15 (55.6)
Yes	19 (19.6)	12 (44.4)
Endoscopic timing			0.106
≤24	73 (75.3)	16 (59.3)
>24	24 (24.7)	11 (40.7)
*H. Pylori* Infection			0.024
Negative	34 (35.1)	3 (11.1)
Positive	63 (64.9)	24 (88.9)
Haemoglobin level			0.018
>8	32 (33)	9 (33.3)
<8	65 (67)	18 (66.7)
Forest classification			0.005
Forest 1	3 (3.1)	7 (26)	0.001
Forest 2	15 (15.5)	9 (33)	0.035
Forest 3	79 (81.4)	11 (41)	0.113
Thrombocytopenia			0.001
No	86 (88.7)	10 (37)	0.324
Yes	11 (11.3)	17 (63)	0.004
Location of ulcer
Gastric ulcer	19 (20)	11 (40.7)	0.254
Duodenal ulcer	78 (80)	16 (59.2)	0.093
Treatment modality
PPI therapy	83 (85.6)	16 (59.3)	0.119
Triple therapy	11 (11.3)	9 (33.3)	0.048
Surgical intervention	3 (3.1)	2 (7.4)	0.192

*p-value stated by using Fisher’s exact test

Table 4: Predictors of rebleeding among 27 study patients

Variables	Univariate		Multivariate
Variables	COR (95% CI)	*p-value	AOR (95% CI)	*p-value
Age, in years
≤60	Ref		Ref
>60	3.22 (1.24-8.36)	0.017	2.47 (0.75-8.22)	0.139
Cigarette smoking
No	Ref		Ref
Yes	1.35 (0.44-4.15)	0.603
Endoscopic timing
≤24 Hours	Ref		Ref
>24 Hours	2.09 (0.85-4.12)	0.106	1.13 (0.98-2.02)	0.239
*H. Pylori* Infection
Negative	Ref		Ref
Positive	1.32 (0.91-3.05)	0.194	1.17 (0.86-2.71)	0.321
Haemoglobin level
>8	Ref		Ref
<8	1.91 (1.13-3.38)	0.046	2.24 (1.39-2.98)	0.002
Forest classification
Forest 1
No	Ref		Ref
Yes	1.76 (1.14-3.29)	0.048	2.51 (1.34-4.02)	0.031
Forest 2
No	Ref		Ref
Yes	1.21 (1.03-2.25)	0.003	1.81 (1.18-2.82)	0.044
Forest 3
No	Ref		Ref
Yes	0.92 (0.12-1.51)	0.162	0.81 (0.67-1.43)	0.702
Thrombocytopenia
No	Ref		Ref
Yes	1.92 (1.14-3.08)	0.041	1.43 (1.15-2.71)	0.011
Ulcer locality (duodenal ulcer)
No	Ref		Ref
Yes	1.09 (0.85-1.57)	0.097	0.86 (0.72-1.48)	0.118
Treatment modality
Medical therapy
No	Ref		Ref
Yes	0.59 (0.61-1.04)	0.053	0.96 (0.85-1.31)	0.094
Surgical therapy
No	Ref		Ref
Yes	0.92 (0.73-1.27)	0.123	0.57 (0.42-1.14)	0.072

Abbreviations: COR-crude odds ratio; AOR-adjusted odds ratio

Furthermore, the re-bleeding rate is a significant issue indicating attentive follow-up care and potential further intervention. Only 2.4% (3/124) of NVUGIB patients were admitted to ICU within a follow-up duration of 60 days, indicating that most cases of NVUGIB are not severe enough to require ICU, however, those admitted to ICU show the importance of having ICU resources available.

Clinical profile of 27 study patients with rebleeding

Among 27 patients, rebleeding was observed among younger age with H. pylori infections (88.9% versus 11.1%, p = 0.024) with low Haemoglobin levels (66.7% versus 33.3%, p = 0.018) Table 3. Furthermore, most of the patients with rebleeding were having thrombocytopenia (63% versus 37%, p = 0.004) and forest classes 1 and 2. Additionally, the incidence of rebleeding was present in 21.8% (27/124).

Predictors of rebleeding among 27 patients with non-variceal upper gastrointestinal bleeding

In our study, rebleeding was observed to be significantly high compared to death and ICU admissions as outcomes of NVUGIB. A total of twelve (12) variables were identified as predictors of rebleeding within 60-day follow-up following univariate analysis.

Thereafter, all variables fitted for a p-value of < 0.2 were exposed to multivariate analysis. After backward eliminations, three (3) variables were found to be statistically significant with a p-value of < 0.05.

Predictors of rebleeding included low Haemoglobin level (AOR 2.24; 95% CI, 1.39-2.98, p = 0.002); Forest class 1 (AOR 2.51; 95% CI, 1.34-4.02, p = 0.031); Forest class 2 (AOR 1.81; 95% CI, 1.18-2.82, p = 0.044) and thrombocytopenia (AOR 1.43; 95% CI, 1.15-2.71, p = 0.011) Table 4.

This study aimed to determine the endoscopic findings, outcomes, and predictors of rebleeding among patients with NVUGIB attending BMH, Dodoma, Tanzania.

From the index study, the most prominent endoscopic finding was gastritis accounting for 42.01% and the least was Eosophageal tumor accounting for 1.78% Figure 3. This is quite different from the study conducted in Pakistan that showed the most prominent endoscopic finding was peptic ulcer accounting for 34% followed by erosive gastritis (32%) and the least was duodenal telangiectasia accounting for and 1% [23]. Another study showed that the leading endoscopic finding was a duodenal ulcer [24]. Also, another study conducted in Saudi Arabia showed that the leading endoscopic finding was esophagitis accounting for 38% of all endoscopic findings in patients with NVUGIB [25]. This variation could be explained by the differences in the clinical profile of the study participants and epidemiological trends of infectious causes such as Helicobacter pylori. Results from the index study show that Oesophageal tumor accounting for 1.78% was similar to the findings in the study done by Hussain et al., [23] whereby they also found that oesophageal tumours accounted for 2%.

Furthermore, the index study showed that the outcomes of death, rebleeding, and ICU admission accounted for 4%, 21.8%, and 2.4% respectively Table 2. While, a study conducted in Asia showed that death, rebleeding, and ICU admission accounted for 2.5%, 10.4%, and 4.4% respectively [26]. The results from the index study are quite lower compared to the study conducted in Turkey whereby the 90-day follow-up showed that the outcome of NVUGIB was death 16% and rebleeding 10% [27]. This difference or variation can be postulated to be due to the index study recruited in patients and outpatients while the study by Durak et al., [27] only recruited patients with a follow-up of 90 days, and these patients were seriously sick compared to most of the outpatients involved in our study.

Also, from the index study, the Forrest class 1 and 2 was significantly associated with rebleeding within the 60days follow-up. This is similar to the previous studies done whereby the Forrest class 1 was also associated with rebleeding [28, 29]. This similarity could be explained by the fact that blood vessels in Forrest class 1 and 2 are likely to bleed.

From the index study, the Haemoglobin level was significantly associated with rebleeding within the 60 days follow-up whereby, those with Haemoglobin levels below 8g/dl had higher odds of rebleeding. This is similar to the previous studies done whereby those with low Haemoglobin levels were associated with rebleeding [28, 30, 31]. This similarity could be explained by the reason that those with low haemoglobin have had chronic blood loss that led to iron deficiency.

Furthermore, from the index study thrombocytopenia was significantly associated with rebleeding within the 60 days follow-up period. This is similar to the previous studies done whereby those with low levels of platelets were associated with rebleeding [28, 30, 31]

Limitations of the study

Our study had several limitations. First, the small sample size could limit the generalizability of the findings, making it difficult to draw strong conclusions applicable to the broader population. Second, the study was conducted at a single hospital Centre limits the external validity of the findings, as the patient population and healthcare resources at this hospital may not be representative of other regions in Tanzania or beyond. Third, limited or lack of follow-up data may hinder the ability to assess long-term outcomes, which is crucial for evaluating the effectiveness of treatments or interventions. In our setting, limitations in access to advanced diagnostic tools, therapeutic interventions, and comprehensive patient management, potentially affect the study outcomes. Fourth, most of the patients in our settings particularly those with upper gastrointestinal bleeding they do not sick medication they are only sick when they’re in danger of their lives this small sample represents only insured patient

Non-variceal upper gastrointestinal bleeding is common in Dodoma and the most common endoscopic finding was Gastritis whereas, the esophageal tumor was the least common. Rebleeding is a common outcome among patients with non-variceal upper gastrointestinal bleeding and is associated with thrombocytopenia, low Haemoglobin level, and forest classes. Therefore, intensive management is crucial to improve patient outcomes. However, future research is warranted.

AOR Adjusted Odds Ratio

BLR Binary Logistic Regression

BMC Bugando Medical Centre

BMH Benjamin Mkapa Hospital

BP Blood Pressure

CI Confidence Interval

COR Crude Odds Ratio

DM Diabetes Mellitus

EGD Esophagogastroduodenoscopy

FBP Full blood Picture

GERD Gastroesophageal Reflux Disease

GI Gastrointestinal

IBM-SPSS Internal Business Machines-Statistical Package for Social Sciences

ICU Intensive Care Unit

IEE Image-Enhanced Endoscopy

INR International Normalized Ratio

KCMC Kilimanjaro Christian Medical Centre

LFT Liver Function Test

NSAIDs Non-Steroidal Anti-Inflammatory Drugs

NVUGIB Non-Variceal Upper Gastrointestinal Bleeding

PI Principal investigator

PR Pulse Rate

PUB Peptic Ulcer Bleeding

PUD Peptic Ulcer Disease

RAs Research Assistants

RFT Renal Function Test

RR Respiratory Rate

RUT Rapid Urease Test

SAT Stool Antigen Test

SBT Serology Blood Tests

TZS Tanzania Shillings

UGIB Upper Gastrointestinal Bleeding

VUGIB Variceal Upper Gastrointestinal Bleeding

WLI White Light Imaging

Ethical Consideration

The ethical Permission to conduct this study was granted by the University of Dodoma's research and publication committee, and data collection was approved by Benjamin Mkapa Hospital under reference numbers MA.84/261/02/A/66/8 and AB.150.293.01/"A"/99, respectively. Each study participant was individually and privately informed on the aim and purpose of the study, risks, and benefits of participating in the study. All participants accepted to participate in the study and signed a written informed consent form. Prepared unique identification numbers were used to maintain the patient’s confidentiality and data will be accessible by the investigator and research assistant. Additionally, all patients who needed further care were linked to a specific area for interventions

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Funding

Not applicable

Authors’ contributions

SM: Conceptualization, data curation, formal analysis, investigation, methodology, formal analysis, writing-original draft;

BA: supervision, conceptualization, methodology, investigation, writing review and editing;

ES: supervision, conceptualization, methodology, investigation, writing review and editing;

Acknowledgments

Not applicable

Availability of data and materials

The datasets generated and/or analyzed during the current study are not publicly available due to ethical issues with the patients (considering privacy and confidentiality) but are available from the corresponding author upon reasonable request.

Kwon JH, Kim JS. Transcatheter arterial embolization of acute nonvariceal upper gastrointestinal bleeding refractory to endoscopic hemostasis. Hong Kong J Radiol. 2020;23:164–75.
Bitar SM, Moussa M. The risk factors for the recurrent upper gastrointestinal hemorrhage among acute peptic ulcer disease patients in Syria: A prospective cohort study. Ann Med Surg. 2022;74 December 2021:103252.
Cañamares-Orbís P, Lanas Arbeloa Á. New trends and advances in non-variceal gastrointestinal bleeding—series II. J Clin Med. 2021;10.
Romstad KK, Detlie TE, Søberg T, Thomas O, Ricanek P, Jahnsen ME, et al. Treatment and outcome of gastrointestinal bleeding due to peptic ulcers and erosions–(BLUE study). Scand J Gastroenterol. 2022;57:8–15.
Almadi MA, Almutairdi A, Alruzug IM, Aldarsouny TA, Semaan T. Upper gastrointestinal bleeding : Causes and patient outcomes. 2021. https://doi.org/10.4103/sjg.SJG.
Opio CK, Rejani L, Kazibwe F, Ocama P. The diagnostic accuracy of routine clinical findings for detection of esophageal varices in rural sub-saharan africa where schistosomiasis is endemic. Afr Health Sci. 2019;19:3225–34.
Bazarah SM, Alotaibi RM, Alghamdi RA, Waheeb AS, Rafeea WA, Talab SK, et al. Identifying the Demographic, Clinical, and Endoscopic Findings of Gastroesophageal Reflux Disease in Patients With Helicobacter pylori Infection at King Abdulaziz University Hospital, Jeddah, Saudi Arabia. Cureus. 2022;14:1–7.
Forgerini M, Urbano G, De Nadai TR, Zapata-Cachafeiro M, Kemp R, Mastroianni PDC. Epidemiological profile of patients with non-variceal upper gastrointestinal bleeding secondary to peptic disease in a tertiary referral brazilian hospital. Arq Gastroenterol. 2021;58:202–9.
Mirghani HO, Khamees H. Predictors of Bleeding of Esophageal Varices Among Patients with Portal Hypertension-Sudan. 2016;4:103–6.
On A, Do M, Tr O. Endoscopic findings in upper gastrointestinal bleeding patients at Lacor hospital , northern Uganda. Afr Health Sci. 2012;12:2010–3.
Kiringa SK, Quinlan J, Ocama P, Mutyaba I, Kagimu M. Prevalence, short term outcome and factors associated with survival in patients suffering from upper gastrointestinal bleeding in a resource limited-setting, the case of mulago hospital in Kampala, Uganda. Afr Health Sci. 2020;20:426–36.
Manko M, Bello A, Mohammed M, Egbegbedia P, Daniyan M, Jabir A, et al. Demographic Profile and Endoscopic Findings among Patients with Upper Gastrointestinal Bleeding in Ahmadu Bello University Teaching Hospital, Zaria, North‑Western Nigeria. 2020;:2020–3.
Chofle AA, Jaka H, Koy M, Smart LR, Kabangila R, Ewings FM, et al. mortality among patients admitted with haematemesis in Mwanza , Tanzania : a prospective cohort study. 2014;:1–10.
Jaka H, Koy M, Liwa A, Kabangila R, Mirambo M, Scheppach W, et al. A fibreoptic endoscopic study of upper gastrointestinal bleeding at Bugando Medical Centre in northwestern Tanzania: a retrospective review of 240 cases. BMC Res Notes. 2012;5:200.
Suba M, Ayana SM, Mtabho CM, Kibiki GS. SHORT COMMUNICATION The Aetiology , Management and clinical Outcome of Upper Gastrointestinal Bleeding among Patients Admitted at the Kilimanjaro Christian Medical Centre in Moshi , Tanzania. 2010;12:304–8.
Asotibe JC, Shaka H, Akuna E, Shekar N, Shah H, Ramirez M, et al. Outcomes of Non-Variceal Upper Gastrointestinal Bleed Stratified by Hospital Teaching Status: Insights From the National Inpatient Sample. Gastroenterol Res. 2021;14:268–74.
Elfert K, Love J, Elromisy E, Jaber F, Nayudu S, Ho S, et al. Clinical outcomes of nonvariceal upper gastrointestinal bleeding in nonagenarians and octogenarians: a comparative nationwide analysis. Clin Endosc. 2024;57:342–9.
Falcão D, Alves da Silva J, Pereira Guedes T, Garrido M, Novo I, Pedroto I. The Current Portrayal of Non-Variceal Upper Gastrointestinal Bleeding in a Portuguese Tertiary Center. GE Port J Gastroenterol. 2021;28:392–7.
Ghazali WSW, Zainudin WMKBW, Yahya NK, Ismail AM, Wong KK. Older age and diclofenac are associated with increased risk of upper gastrointestinal bleeding in gout patients. PeerJ. 2021;9:1–12.
Moledina SM, Komba E. Risk factors for mortality among patients admitted with upper gastrointestinal bleeding at a tertiary hospital: A prospective cohort study. BMC Gastroenterol. 2017;17:1–11.
Mullady DK, Wang AY, Waschke KA. AGA Clinical Practice Update on Endoscopic Therapies for Non-Variceal Upper Gastrointestinal Bleeding: Expert Review. Gastroenterology. 2020;159:1120–8.
Alali AA, Barkun AN. An update on the management of non-variceal upper gastrointestinal bleeding. Gastroenterol Rep. 2023;11 January:1–18.
Hussain MA, Durrani MA, Chaudry JZ, Iqbal S. Etiological Pattern and Outcome of Non-Variceal Upper G.I. Bleeding. Gomal J Med Sci. 2016;14:41–5.
Mahajan P, Chandail VS. Etiological and Endoscopic Profile of Middle Aged and Elderly Patients with Upper Gastrointestinal Bleeding in a Tertiary Care Hospital in North India: A Retrospective Analysis. J Midlife Health. 2017;8:137–41.
Almadi MA, Almutairdi A, Alruzug IM, Aldarsouny TA, Semaan T, Aldaher MK, et al. Upper gastrointestinal bleeding: Causes and patient outcomes. Saudi J Gastroenterol. 2021;27:20–7.
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.
Durak MB, Şimşek C, Yüksel İ. Clinical Outcomes of Older Patients with Non-Variceal Upper Gastrointestinal Bleeding Taking Anti-Thrombotic or Non-Steroidal Anti-Inflammatory Agents. Turkish J Gastroenterol. 2023;34:918–24.
Nam JH, Jeon TJ, Cho JH, Kim JH. Risk factors of the rebleeding according to the patterns of nonvariceal upper gastrointestinal bleeding. Turkish J Gastroenterol. 2017;28:266–71.
Kim WS, Kim SH, Joo MK, Park JJ, Lee BJ, Chun HJ. Re-bleeding and all-cause mortality risk in non-variceal upper gastrointestinal bleeding: focusing on patients receiving oral anticoagulant therapy. Ann Med. 2023;55.
Hajiagha Mohammadi AA, Reza Azizi M. Prognostic factors in patients with active non-variceal upper gastrointestinal bleeding. Arab J Gastroenterol. 2019;20:23–7.
Yoo AY, Joo MK, Park JJ, Lee BJ, Kim SH, Kim WS, et al. Recurrent Non-Variceal Upper Gastrointestinal Bleeding among Patients Receiving Dual Antiplatelet Therapy. Diagnostics. 2023;13:1–10.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Endoscopic findings and predictors of outcome among patients with non-variceal upper gastrointestinal bleeding attending Benjamin Mkapa Hospital in Dodoma, Tanzania

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Methods

Study Design

Study setting

Study Population

Sample Size

Sampling Technique

Data collection tools and methods

Study procedure

Endoscopic Procedures

Laboratory procedure and data collection

Study variable and definition:

Independent variables:

Social Demographic Characteristics

Clinical characteristics:

Laboratory tests

Intermediate variable

Data Processing and Analysis

Results

Baseline and clinical characteristics of the study population

Treatment modality of NVUGIB of 124 study patients

Outcomes of Patients with non-variceal upper gastrointestinal bleeding

Discussion

Conclusion

Abbreviations

Declarations

Consent for publication

Competing interests

Funding

Acknowledgments

References

Additional Declarations

Supplementary Files

Status:

Version 1