|Year : 2021 | Volume
| Issue : 8 | Page : 1211-1216
Early Endoscopy Decrease the Length of Hospital Stay and the Costs in Patients with Upper Gastrointestinal Bleeding
E Arslan1, E Ozakin1, T Temel2, AH Ozakyol2, N Acar1, FB Kaya1, ME Canakci1, T Caglayan1
1 Department of Emergency Medicine, Eskisehir Osmangazi University Medical Center, Meselik, Eskisehir, Turkey
2 Department of Gastroenterology, Eskisehir Osmangazi University Medical Center, Meselik, Eskisehir, Turkey
|Date of Submission||07-Jul-2020|
|Date of Acceptance||12-Jan-2021|
|Date of Web Publication||14-Aug-2021|
Dr. E Ozakin
Department of Emergency Medicine, Eskisehir Osmangazi University Medical Center, Meselik - 26480, Eskisehir
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: Acute gastrointestinal (GI) bleeding is a common cause of hospitalization. There are conflicting results regarding the effectiveness of early endoscopy in patients with upper GI bleeding. Objective: The study aimed to determine the benefit of early endoscopy and the epidemiological characteristics of patients presenting to the emergency department with non-variceal acute upper GI bleeding. Methods: Patients over 18 years of age who presented to the emergency department with upper GI between 2015 and 2016 and underwent endoscopy were included in the study. The patients were divided into two groups: early endoscopy group (endoscopy within the 24 h) and late endoscopy group (endoscopy after 24 h). Results: Of 104 patients, 57.7% were man, and the mean age was 66.27 ± 17.64 years. Of the patients who underwent endoscopy, 80 (76.9%) were in the early endoscopy group. There was no difference in blood transfusion needs (P = 0.388), re-bleeding (P = 0.137), the need for surgery, and mortality rates with regard to the timing of endoscopy. The results of the receiver-operating characteristics curve analysis revealed patients with a GBS ≥ 9 were taken as high-risk, and a prognosis analysis was performed accordingly. However, in patients undergoing early endoscopy, a difference was detected with respect to the length of hospital stay (P = 0.011) and treatment costs (P = 0.030). In the comparison with the admission time (working/non-working h) and variables, there were no significant differences in the length of hospital stay (P = 0.230), transfusion needs (P = 0.348), re-bleeding frequency (P = 0.905), and treatment costs (P = 0.094). Conclusion: Endoscopy within 24 h in the setting of acute upper non-variceal GI bleeding is associated with an increase in the length of hospital stay and treatment costs, but is not associated with re-bleeding, transfusion needs, need for surgery, and mortality.
Keywords: Early endoscopy, length of hospital stay, treatment costs
|How to cite this article:|
Arslan E, Ozakin E, Temel T, Ozakyol A H, Acar N, Kaya F B, Canakci M E, Caglayan T. Early Endoscopy Decrease the Length of Hospital Stay and the Costs in Patients with Upper Gastrointestinal Bleeding. Niger J Clin Pract 2021;24:1211-6
|How to cite this URL:|
Arslan E, Ozakin E, Temel T, Ozakyol A H, Acar N, Kaya F B, Canakci M E, Caglayan T. Early Endoscopy Decrease the Length of Hospital Stay and the Costs in Patients with Upper Gastrointestinal Bleeding. Niger J Clin Pract [serial online] 2021 [cited 2022 Jan 25];24:1211-6. Available from: https://www.njcponline.com/text.asp?2021/24/8/1211/323855
| Introduction|| |
Despite recent improvements in diagnostic and endoscopic tools and new pharmacological agents in the treatment of peptic ulcer, upper gastrointestinal (GI) bleeding might reach a mortality rate of 10% and is still critical for emergency medical care.
Acute GI bleeding is a common cause of hospitalization, leading to approximately 400.000 hospital admissions per year and mortality rates of 5-10%. The severity covers a broad spectrum from subclinical occult bleeding to abundant bleeding, and from chronic anemia to acute hypovolemic shock.
Early endoscopy reduces the recurrence of upper GI bleeding and is also recommended as a first-line treatment. Numerous randomized controlled clinical studies and meta-analyses suggest that early endoscopic hemostatic treatment decreases the frequency of re-bleeding, need for immediate surgical intervention and mortality in patients with acute non-variceal upper GI bleeding. In short, in cases with acute upper GI bleeding, an endoscopic examination is critical for establishing an accurate diagnosis and obtaining successful treatment results.,,
Whereas prior studies supported early therapeutic endoscopy in patients with upper GI bleeding, recent studies and some opinions have revealed disagreements in endoscopic treatment. Therapeutic endoscopy might be unsafe due to its complications, such as perforation, hypertension, aspiration pneumonia, re-bleeding, ventricular arrhythmia, and myocardial ischemia. Besides, many studies question the cost-effectiveness of the treatment suggesting that early endoscopic procedure does not affect the outcome and 80% of the patients responded to the conservative treatment. Also, there are contradictory results about the efficacy of urgent endoscopy performed in patients with upper GI bleeding.,
Considering the contradictory results about the efficacy of early therapeutic endoscopy performed in patients with upper GI bleeding, this study aims to assess the benefit of early endoscopy.
The primary outcome of this study was to investigate the effect of the time to endoscopy from the emergency department admission on mortality. The secondary outcome was to investigate the effect of this period on re-bleeding, requirement for surgery, length of hospital stay and cost.
| Methods|| |
This study was reviewed and approved by the Research Ethics Committee of The College of Medicine of XXX University (Reference No: 80558721/320).
Study design and setting
This study included patients over 18 years of age who presented to the Emergency Department (ED) of the University Hospital with upper GI bleeding and had therapeutic upper GI endoscopy. This is a retrospective analysis of the data prospectively collected between 1 January 2015 and 31 December 2015. Patients with conditions such as peritonitis, perforation, acute myocardial infarction, major abdominal aneurysm, and patients having variceal hemorrhage after the endoscopy were excluded from the study as they would have an impact on a number of prognostic factors, the course of the treatment, and change the overall approach.
Endoscopy time was calculated as the time from admission to the ED to GI endoscopy onset time. Based on previous studies, the patients were divided into two groups, and those who underwent endoscopy within 24 h after the ED admission were identified as early endoscopy group whereas those who underwent endoscopy after 24 h were identified as late endoscopy group.,
A standardized data form was used for patients' data such as demographic characteristics, comorbid diseases, habits, medication, history of bleeding, vital signs, laboratory findings and blood transfusion needs.
After achieving hemodynamic stabilization, endoscopic procedures were performed either in the department patients referred to, an intensive care unit, or an outpatient endoscopy unit. Endoscopic procedures were carried out by the academicians working at the Department of Gastroenterology of XXX University. Endoscopic interventions included injection therapy (adrenaline, saline, alcohol), bipolar electrocoagulation, argon plasma coagulation and hemo-clips.
Risk analysis was performed using Glasgow Blatchford Score (GBS) and the Rockall scoring system,, and the Forrest classification was used to classify peptic ulcer lesions. Following the procedure, the length of hospital stays, additional therapies, re-bleeding, need for surgical intervention, mortality and treatment costs were recorded.
Quantitative data which followed a normal distribution were expressed as means ± standard deviation and those that did not follow a normal distribution were presented as median and percentages. Qualitative data were not represented by percentages. Shapiro Wilk's test was used to verify data compatibility with a normal distribution. For the comparison of the groups which are not compatible with a normal distribution, Mann-Whitney U test was used when the number of groups was two, and Kruskal-Wallis H test was used when the number of groups was three or more. Spearman correlation analysis was used to assess the direction and the strength of the relationship between the variables that did not follow the normal distribution. Pearson Chi-Square, Pearson Exact, and Fisher exact analyzes were used for the analysis of the cross tables. ROC (receiver-operating characteristics) analyses were used to determine the appropriate cutoff point for independent markers and to calculate sensitivity and specificity values. IBM SPSS 21.0 and MedCalc 16.8 were used in the analysis. A P value of 5% or lower was set as a statistical significance criterion.
| Results|| |
One hundred four patients that were diagnosed with upper GI bleeding and had upper endoscopy were included in the study. Sixty (57%) of them were man. The mean age of the patients was 66.27 ± 17.64 (min: 21–max: 95). The analysis on the distribution of the endoscopy timing revealed that the procedure was performed approximately in 19.60 ± 18.51 h following the hospital admission. Of the patients who underwent endoscopy, 80 (76.9%) of them were in the early group and 24 (23.1%) of them were in the late group. The underlying causes and demographic characteristics of the study group are presented in [Table 1] and [Table 2].
|Table 2: Demographics and Characteristics data on patients with upper GI bleeding|
Click here to view
Of the 57 transfused patients, 42 (52.5%) underwent endoscopy within the first 24 h and 15 (62.5%) after 24 h. There was no statistically significant difference in terms of transfusion needs according to endoscopic intervention time (P = 0.388).
After the endoscopic interventions, re-bleeding occurred in six patients. Three (3.8%) of them were in the early period and the remaining three (12.5%) were in the late endoscopy group (P = 0.137). There was no statistically significant difference in terms of re-bleeding according to endoscopic intervention time (P = 0.137). The highest transfusion requirement of the lesions according to the Forrest classification was found in class 3 and re-bleeding in classes 2C and 3.
However, a significant decrease was seen in the length of hospital stay and the treatment costs of the patients undergoing early endoscopy (P = 0.011 and P = 0.030, respectively) [Table 3].
Nevertheless, surgery intervention and mortality were not observed in any of these patients.
When identifying patients in the risk group, ROC analyses were used to estimate sensitivity and specificity values and to determine the best cutoff point for independent markers. The result of the ROC curve analysis revealed a GBS more than or equal to 9, and a sensitivity of 70.8% and a specificity of 53.8% were calculated. Thus, patients with a GBS more than nine were taken as high-risk, and prognosis analysis was made accordingly. When the timing of endoscopy and risk scores of the patients were compared, no significant difference was found between patients undergoing early endoscopy and late endoscopy (P = 0.081).
With regard to the patients who were classified as high-risk (GBS ≥ 9) and low-risk (GBS < 9) according to their GBS, we found that patients' risk levels had no effect on the timing of endoscopy and there was no statistically significant difference (P = 0.211). According to the assessment of 60 high-risk patients, the length of hospital stay, the need for blood transfusion, and the treatment costs for the patients undergoing late endoscopy were significantly higher than those for the patients receiving early endoscopy (P = 0.033, P < 0.001, P < 0.001).
It was found that the admission time of the patients (working/non-working hours) did not cause any delay in endoscopy. Patients admitted during non-working hours even underwent endoscopy earlier although there was not a significant difference. When the patients' admission time and prognosis were compared, there was no significant difference in the length of hospital stay, transfusion needs, re-bleeding frequency, and treatment costs. Even though it was not statistically and significantly different, an increase was seen in re-bleeding frequency, transfusion needs, and treatment costs in the patients admitted during non-working hours.
Forrest classification and endoscopic procedures according to early and late endoscopy time are given in [Table 4].
| Discussion|| |
Patients with acute GI bleeding may present to the emergency department with complaints ranging from subclinical occult bleeding to abundant bleeding, from chronic anemia to severe hypovolemic shock. Endoscopic procedure performed to treat upper GI bleeding is not only essential for making accurate diagnoses and achieving successful results but also its application at an early stage of the treatment is critical for reducing the recurrence rate of bleeding. It is also recommended as a first-line treatment.,,
The incidence of upper GI bleeding is two times higher in male patients in all age groups; however, mortality rates for both male and female patients were the same. In the literature, prior studies showed a higher male/female ratio of upper GI bleeding compared to our study.,, We believe that higher rates of additional diseases, primarily liver cirrhosis, and diseases requiring intake of aspirin and other NSAIDs play a significant role in the high prevalence of upper GI bleeding among elderly patients. The mean age in our study was similar to the values in other studies. In cases with upper GI bleeding, one of the most critical factors affecting prognosis, also serving as an etiological element, is the presence of an additional disease. Similar to the current studies, the frequency of additional diseases in upper GI bleeding was also high in our study.,
Furthermore, our study found that the presence of an additional disease caused a significant increase in the length of hospital stay. However, no significant difference was detected in blood transfusion needs, re-bleeding frequency, the need for surgical intervention, and mortality. We think that this resulted from absence of high-risk lesions, a small number of patients with re-bleeding, and the medical therapy performed in the early stages. Similar to the results of the current studies in the literature, our study found that the risk of bleeding in lesions increased as the ulcer incidence rate decreased. Nevertheless, unlike our results, current studies showed that patients with high-risk lesions had higher rates of re-bleeding and mortality. We think that it might result from a small number of patients with high-risk lesions in the sample group.
Similar to the results (5.9%) of current studies in the literature, our study demonstrated that the frequency of therapeutic endoscopic procedure was at lower rates. We think that the lower frequency of therapeutic endoscopic procedure has resulted from the lesions which were unsuitable for endoscopic procedures or difficult to be scanned. Recent studies stated that endoscopic procedures achieved desired results in establishing hemostasis control and improved the prognosis in cases with upper GI bleeding. When evaluating the impact of endoscopy on prognosis, we found no statistically significant difference in re-bleeding frequency and transfusion needs in patients undergoing endoscopy. In our opinion, this was due to the absence of high-risk lesions for re-bleeding according to the Forrest classification (Forrest 2B- 7 patients, 2A- 1 patient, 1B- 2 patients, 3- 1 patient).
Endoscopy is crucial for establishing an accurate diagnosis and obtaining excellent results in patients with upper GI bleeding. In the literature, studies are showing that early endoscopy performed in 24 h shortens the length of hospital stay, reduces the frequency of re-bleeding, and the need for immediate surgical intervention in patients with acute upper GI bleeding. A retrospective research study consisting of more than 30.000 cases found that mortality rates among patients undergoing early endoscopy were two times lower (5.2% versus 11.1%). In the 2010 International Consensus Conference, it was expressed that early endoscopy was convenient for patients with non-variceal upper GI bleeding and it made a significant contribution to prognosis. However, therapeutic endoscopy may not only cause perforation and a more severe degree of bleeding, but it can also increase the treatment cost. Many studies have questioned the cost-efficiency of the treatment as early endoscopy does not affect the outcome and 80% of patients already respond to conservative treatment methods. When examining the effect of early endoscopy on prognosis, we observed that there was no association between early endoscopy and mortality rates, transfusion needs, re-bleeding frequency, and the need for surgical intervention, however, at a statistically significant level, the length of hospital stay and the treatment costs were lower. Therefore, our results showed that the cost-efficiency of endoscopic procedures was high. We think that this resulted from a significant number of lesions that did not require any therapeutic endoscopy and had a low risk for re-bleeding. Therefore, we could not carry out an objective assessment of the impact of early endoscopy on prognosis, for our study mostly consisted of low-risk lesions for re-bleeding.
GBS is considered the most effective method for early risk assessment before performing endoscopy in upper GI bleeding. Current studies have demonstrated that GBS is more effective than Rockall Score in determining the need for blood transfusion, estimating mortality and the need for endoscopic or surgical therapy. Thus, in our study, we used GBS to detect high-risk patients and compared prognosis and the timing of endoscopy in all patients. In our study, the efficiency of The Glasgow-Blatchford Bleeding Score in making an accurate risk assessment, establishing a prognosis, and determining the need for therapy was found to be similar to the results of current studies in the literature. When examining data on prognosis and admission time of the patients, we found that the admission time of the patients (working/non-working hours) did not cause any delay in endoscopy. Patients admitted during non-working hours even underwent endoscopy earlier although there was not a significant difference. Even though there was no statistically significant difference, an increase was detected in the re-bleeding rates, the need for transfusion and the treatment costs in patients admitted during non-working hours. This result reveals that patients admitted during working hours are managed more efficiently and their treatment plans are made more adequately.
This study included only the patients living in a central Anatolian city (Turkey) and was conducted in a single center. That our hospital was the only third-line hospital in the city and patients referred to our hospital had critical conditions with high comorbidity brought about a sample group with a very small number of patients and did not allow us to make a comparison with the general population. Patients with a history of variceal hemorrhage or patients diagnosed with variceal bleeding following the endoscopic procedure were excluded from the study as they would affect numerous prognostic factors, the course of the treatment, and would alter the overall approach. That the most of the lesions detected by endoscopy had a lower risk for bleeding and were unsuitable for therapeutic endoscopy decreased the number of endoscopic procedures performed, thus the impact of early endoscopy on prognosis could not be evaluated objectively. Moreover, parameters of prognosis such as re-bleeding and mortality were assessed only during the period of hospital stay and not followed afterwards.
| Conclusion|| |
Endoscopy within the 24 h in the setting of acute upper non-variceal GI bleeding is associated with length of hospital stay and treatment costs, but is not associated with re-bleeding, transfusion need, need for surgery and mortality. According to our study, further studies are needed for establishing the necessity of emergency endoscopy in patients with non-variceal upper GI bleeding admitted to emergency department.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Nadler M, Eliakim R. The role of capsule endoscopy in acute gastrointestinal bleeding. Therap Adv Gastroenterol 2014;7:87-92.
Barkun A, Bardou M, Marshall JK; Nonvariceal Upper GI Bleeding Consensus Conference Group. Consensus recommendations for managing patients with nonvariceal upper gastrointestinal bleeding. Ann Intern Med 2003;139:843-57.
Pang SH, Ching JY, Lau JY, Sung JJ, Graham DY, Chan FK. Comparing the Blatchford and pre-endoscopic Rockall score in predicting the need for endoscopic therapy in patients with upper GI hemorrhage. Gastrointest Endosc 2010;71:1134-40.
Cook DJ, Guyatt GH, Salena BJ, Laine LA. Endoscopic therapy for acute nonvariceal upper gastrointestinal hemorrhage: A meta-analysis. Gastroenterology 1992;102:139-48.
Vallon AG, Cotton PB, Laurence BH, Armengol Miro JR, Salord Oses JC. Randomised trial of endoscopic argon laser photocoagulation in bleeding peptic ulcers. Gut 1981;22:228-33.
Adler DG, Leighton JA, Davila RE, Hirota WK, Jacobson BC, Qureshi WA, et al
. ASGE guideline: The role of endoscopy in acute non-variceal upper-GI hemorrhage. Gastrointest Endosc 2004;60:497-504.
Ramirez F, Cifuentes C, Mavares J, Voso J, Monasterios W, Hinestrosa H, et al
. [Upper gastrointestinal hemorrhage. Prospective analysis of 741 cases]. G E N 1993;47:139-44.
Sarin N, Monga N, Adams PC. Time to endoscopy and outcomes in upper gastrointestinal bleeding. Can J Gastroenterol 2009;23:489-93.
Cooper GS, Chak A, Way LE, Hammar PJ, Harper DL, Rosenthal GE. Early endoscopy in upper gastrointestinal hemorrhage: Associations with recurrent bleeding, surgery, and length of hospital stay. Gastrointest Endosc 1999;49:145-52.
Rockall TA, Logan RF, Devlin HB, Northfield TC. Selection of patients for early discharge or outpatient care after acute upper gastrointestinal haemorrhage. National Audit of acute upper gastrointestinal haemorrhage. Lancet 1996;347:1138-40.
Tham J, Stanley A. Clinical utility of pre-endoscopy risk scores in upper gastrointestinal bleeding. Expert Rev Gastroenterol Hepatol 2019;13:1161-7.
Tseng PH, Liou JM, Lee YC, Lin LY, Yan-Zhen Liu A, Chang DC, et al
. Emergency endoscopy for upper gastrointestinal bleeding in patients with coronary artery disease. Am J Emerg Med 2009;27:802-9.
Sezgin O, Altintas E, Tombak A. Effects of seasonal variations on acute upper gastrointestinal bleeding and its etiology. Turk J Gastroenterol 2007;18:172-6.
Elmunzer BJ, Young SD, Inadomi JM, Schoenfeld P, Laine L. Systematic review of the predictors of recurrent hemorrhage after endoscopic hemostatic therapy for bleeding peptic ulcers. Am J Gastroenterol 2008;103:2625-32; quiz 33.
Lam KL, Wong JC, Lau JY. Pharmacological treatment in upper gastrointestinal bleeding. Curr Treat Options Gastroenterol 2015;13:369-76.
Theocharis GJ, Thomopoulos KC, Sakellaropoulos G, Katsakoulis E, Nikolopoulou V. Changing trends in the epidemiology and clinical outcome of acute upper gastrointestinal bleeding in a defined geographical area in Greece. J Clin Gastroenterol 2008;42:128-33.
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.
Blatchford O, Murray WR, Blatchford M. A risk score to predict need for treatment for upper-gastrointestinal haemorrhage. Lancet 2000;356:1318-21.
[Table 1], [Table 2], [Table 3], [Table 4]