ABSTRACT
BACKGROUND:
Upper gastrointestinal (GI) bleeding is one of the most common reasons for emergency department (ED) visits. This study aimed to evaluate the predictive power of the AIMS65 and pre-endoscopy Rockall scores in predicting in-hospital mortality in patients that presented to ED and were diagnosed with the upper GI bleeding.
METHODS:
Data of patients aged 18 years and older, who visited ED of Kartal Dr. Lütfi Kırdar City Hospital during the study period and were diagnosed with upper GI bleeding, were obtained from the electronic-based hospital information system and analyzed retrospectively. Each scoring system was compared using the receiver operating characteristic (ROC) curve analysis.
RESULTS:
The study was completed with 592 patients. The mean age of the patients was 63.5±19.0 years, and 68.6% were male. The total in-hospital mortality rate was 5.2%. In the ROC analysis of the AIMS65 and pre-endoscopy Rockall scores in the prediction of in-hospital mortality, the area under the curve values was calculated as 0.822 (95% confidence interval [CI]: 0.788–0.852) and 0.777 (95% CI: 0.741–0.810), respectively. When these two scoring systems were compared, neither had statistically significant superiority over the other in predicting in-hospital mortality.
CONCLUSION:
The AIMS65 and pre-endoscopy Rockall scores can be used to predict in-hospital mortality in patients with GI bleeding. However, since the AIMS65 score consists of only five variables that can easily be calculated in ED, we recommend its use in clinical practice.
Keywords: AIMS-65 score, gastrointestinal bleeding, mortality, rockall score
INTRODUCTION
Upper gastrointestinal (GI) bleeding is one of the most common reasons for emergency department (ED) visits. In the United States, more than half a million patients are admitted to hospitals each year due to GI bleeding.[1,2] Despite new endoscopic techniques and pharmacological treatments, the mortality rate due to this condition can reach 10%.[3] Therefore, many international guidelines on the management of patients with GI bleeding recommend the use of scoring systems to predict undesirable outcomes, such as rebleeding and mortality.[4–6]
The AIMS65 score was derived by Saltzman et al.[7] in 2011 by retrospectively screening the data of 29,222 patients from 187 hospitals and consists of five parameters (albumin less than 3.0 g/dL, international normalized ratio [INR] >1.5, altered mental status, systolic blood pressure 90 mm Hg or lower, and age older than 65 years). This score is used to estimate in-hospital mortality in patients with the upper GI bleeding. The probability of in-hospital mortality in a patient with an AIMS65 score of 0 is 0.3%, while it increases to 24.5% in a patient with a score of 5.
The pre-endoscopy Rockall score was developed in 1996 by Rockall et al.,[8] who prospectively evaluated the data of 3981 patients from 74 hospitals and identified three parameters (age, shock status, and comorbidities). The probability of mortality is 0.2% in a patient with a pre-endoscopy Rockall score of 0 and 50% in a patient with a score of 7.
The present study aimed to evaluate the predictive power of the AIMS65 and pre-endoscopy Rockall scores in predicting in-hospital mortality in patients that presented to ED and were diagnosed with the upper GI bleeding and to examine whether either method is superior to the other.
MATERIALS AND METHODS
This retrospective, observational, and single-center study was conducted at the Emergency Medicine Clinic of Kartal Dr. Lütfi Kırdar City Hospital. The institutional management approved the analysis and issued a waiver of consent (ethics committee ruling number: 2022/514/230/18, date: 27.07.2022).
The data of all patients aged 18 years and older, who visited ED between June 1, 2021, and June 1, 2022, and were diagnosed with upper GI bleeding according to the International Classification of Diseases, 10th revision) diagnostic codes, were obtained by screening the electronic medical records of the hospital. Patients with a diagnosis other than GI bleeding, cases in which the AIMS65 and pre-endoscopy Rockall scores could not be calculated, patients that had been transferred to our ED from another hospital, those that died in or were discharged from ED, and those without follow-up data were excluded from the study.
The first two researchers collected the following data by examining the electronic medical records of each patient: Age, gender, vital parameters, symptoms at admission, comorbidities, medications used, laboratory parameters, and in-hospital mortality status. In addition, each patient’s AIMS65 and pre-endoscopy Rockall scores were calculated separately. After the data set was established, it was reviewed and confirmed by the remaining researchers.
In this study, we aimed to examine the relationship between the AIMS65 and pre-endoscopy Rockall scores and in-hospital mortality among patients with the upper GI bleeding. The primary outcome of the study was the in-hospital mortality rate.
Statistical Analysis
To perform statistical analysis, SPSS v. 25.0 software package (SPSS Inc., Chicago, IL, USA) and MedCalc ver. 12.5 (MedCalc Software Ltd, Ostend, Belgium) was used. Descriptive statistics were presented as mean and standard deviation values and percentage distribution. The conformance of the data to the normal distribution was cross-checked with the Kolmogorov-Smirnov test. The receiver operating characteristic (ROC) curve analysis was used to determine the cutoff values of the AIMS65 and Rockall scores in predicting mortality. The optimal cutoff value, 95% confidence interval (CI), area under the curve (AUC), positive predictive value, and negative predictive value were calculated. The significance level was accepted as p<0.05.
RESULTS
Of the 627 patients diagnosed with the upper GI bleeding according to the ICD-10 codes, 592 were included in the study. Excluded were 35 patients, including 23 whose scores could not be calculated due to missing data, four that died in ED, three that had been transferred from another hospital, and five without follow-up data were excluded from the study. The mean age of the 592 patients was 63.5±19.0 years, of which 68.6% were male. Table 1 presents the main characteristic findings of the patients. To compare their various characteristics, the patients were categorized into two groups: Survivors and non-survivors.
Table 1.
Patient characteristics
| Survivors | Non-survivors | Total | p-value | ||||
|---|---|---|---|---|---|---|---|
|
|
|
|
|||||
| Mean Number | SD Percentage | Mean Number | SD Percentage | Mean Number | SD Percentage | ||
| Gender | 0.037 | ||||||
| Female | 171 | 30.5 | 15 | 48.4 | 186 | 31.4 | |
| Male | 390 | 69.5 | 16 | 51.6 | 406 | 68.6 | |
| Age, years | 62.8 | 19.1 | 77.1 | 12.1 | 63.5 | 19.0 | 0.001 |
| Systolic blood pressure, mmHg | 122.9 | 16.9 | 108.7 | 18.6 | 122.2 | 17.4 | 0.001 |
| Heart rate, bpm | 97.7 | 14.2 | 104.5 | 16.6 | 98.1 | 14.3 | 0.010 |
| Hemoglobin, g/dL | 9.2 | 2.8 | 8.5 | 3.1 | 9.2 | 2.8 | 0.171 |
| Blood urea nitrogen, mg/dL | 79.8 | 51.5 | 127.3 | 66.2 | 82.3 | 53.3 | 0.001 |
| Creatinine, mg/dL | 1.16 | 1.03 | 2.15 | 2.34 | 1.21 | 1.15 | 0.001 |
| Albumin, mg/dL | 35.3 | 6.4 | 28.4 | 6.5 | 34.9 | 6.6 | 0.001 |
| International normalized ratio | 1.06 | 0.25 | 2.10 | 2.59 | 1.12 | 0.67 | 0.001 |
| AIMS65 score | 1.04 | 0.7 | 2.1 | 0.8 | 1.1 | 0.7 | 0.001 |
| Pre-endoscopy Rockall score | 2.7 | 1.7 | 4.3 | 1.1 | 2.8 | 1.7 | 0.001 |
| Steroids | 17 | 3.0 | 3 | 9.7 | 20 | 3.4 | 0.046 |
| Anticoagulants | 81 | 14.4 | 7 | 22.6 | 88 | 14.9 | 0.215 |
| Antiplatelet agents | 151 | 26.9 | 13 | 41.9 | 164 | 27.7 | 0.069 |
| Non-steroidal anti-inflammatory agents | 78 | 13.9 | 11 | 35.5 | 89 | 15.0 | 0.001 |
| Length of hospital stay (days) | 5.8 | 2.5 | 6.5 | 2.9 | 5.9 | 2.5 | 0.156 |
SD: Standard deviation.
The total in-hospital mortality rate was 5.2% (31 patients), and the mean age of these patients was 77.1±12.1 years. The mean AIMS65 and pre-endoscopy Rockall scores, albumin-INR-blood urea nitrogen-creatinine levels, and heart rate were found to be higher in the non-survivors compared to the survivors (p<0.001).
According to the ROC analysis of the AIMS65 score in the prediction of in-hospital mortality, the AUC value was 0.822 (95% CI: 0.788–0.852), the Youden index was 0.578, and the p-value was 0.001 (Table 2 and Fig. 1). The ROC analysis of the pre-endoscopy Rockall score in the prediction of in-hospital mortality revealed an AUC value of 0.777 (95% CI: 0.741–0.810), Youden index of 0.421, and p value of 0.001 (Table 2 and Fig. 1). As a result of the statistical analysis, both scores were found to be statistically significant in predicting in-hospital mortality (p<0.001). When the two scores were compared, neither had statistically significant superiority over the other in the prediction of in-hospital mortality (Table 2).
Table 2.
Optimum cut-off values of the AIMS65 and pre-endoscopy Rockall scores in predicting in-hospital mortality
| AUC | Cut-off | Sensitivity | Specificity | LR+ | LR- | PPV | NPV | Youden index | p-value | |
|---|---|---|---|---|---|---|---|---|---|---|
| AIMS65 | 0.822 | >1 | 80.7 | 77.2 | 3.5 | 0.3 | 16.3 | 98.6 | 0.578 | 0.316 |
| (0.788–0.852) | ||||||||||
| Pre-endoscopy | 0.777 | >3 | 77.4 | 64.7 | 2.2 | 0.4 | 10.8 | 98.1 | 0.421 | |
| Rockall score | (0.741–0.810) |
AUC: Area under the curve; LR: Likelihood ratio; PPV: Positive predictive value; NPV: Negative predictive value.
Figure 1.
Receiver operating characteristic analysis of the AIMS65 and pre-endoscopy Rockall scores in predicting in-hospital mortality.
DISCUSSION
In this study, the predictive power of the AIMS65 and pre-endoscopy Rockall scores in predicting in-hospital mortality in patients that visited ED due to the upper GI bleeding was examined, and it was concluded that both scores were successful, and neither was superior to the other.
Various scoring systems are used to classify high-risk patients and differentiate low-risk patients in the presence of the upper GI bleeding. Scoring systems can assist ED physicians in predicting mortality and rebleeding rates in these patients and guide treatment decisions. However, not every scoring system can be used in every hospital. For example, the Rockall score, which includes endoscopic findings, will not be useful in a hospital that does not have access to endoscopy. Therefore, in the present study, we examined the AIMS65 and pre-endoscopy Rockall scores, which can be easily calculated in almost all EDs.
In the literature, there are many studies examining clinical risk scores in patients with GI bleeding.[9–11] Hyett et al.[12] investigated the AIMS65 score and the Glasgow-Blatchford score (GBS) in patients with the upper GI bleeding and concluded that the former was superior in predicting in-hospital mortality, while the latter was more successful in predicting blood transfusion requirement. In a study from Korea, the AIMS65 score had a higher AUC value compared to GBS, Rockall, and pre-endoscopy Rockall in the estimation of in-hospital mortality (0.84 vs. 0.72, 0.75, and 0.74, respectively), but no statistically significant difference was between these scoring systems.[13] In another study, Robertson et al.[14] evaluated 424 patients with GI bleeding and found the AIMS65 and full Rockall scores to be superior to GBS and the pre-endoscopy Rockall score in predicting in-hospital mortality. In a multicenter and prospective study, the risk scores for 90-day mortality and rebleeding outcomes of 1024 patients with the upper GI bleeding were examined, and the ABC and pre-endoscopy Rockall scores were determined to be superior to the AIMS65 score and GBS in predicting 90-day mortality.[15] In our study, although the AIMS65 score had a higher AUC value than the pre-endoscopy Rockall score (0.82 vs. 0.77), both were similarly successful in predicting in-hospital mortality, with no statistically significant superiority.
In our cohort, the total mortality rate was 5.2%. In the literature, studies conducted with patients with GI bleeding have reported different mortality rates. Yaka et al.[16] determined the in-hospital mortality rate as 7.1%, and Stanley et al.[17] reported the 30-day mortality rate to be 7%, while Robertson et al.[14] calculated the in-hospital mortality rate as 4.2%, and Budimir et al.[18] found the 30-day mortality rate to be 5.2%. In light of these data, the results of our study are in agreement with the literature.
Limitations
There are certain limitations to our study. First concerns the single-center design and a relatively small sample; therefore, our findings should be validated in a larger multicenter cohort. In addition, due to the retrospective nature of our study, data were obtained from the electronic registry system, resulting in incomplete or outdated information. However, to minimize errors, the medical records collected were first recorded in a form by two researchers, and then double-checked by the remaining two other researchers.
Conclusion
In this study, it was concluded that both the AIMS65 and pre-endoscopy Rockall scores were successful in predicting in-hospital mortality in patients with GI bleeding. However, since the AIMS65 score consists of only five variables (albumin, INR, mental status, systolic blood pressure, and age) and can be easily calculated in EDs, we recommend the use of this score in clinical practice.
Footnotes
Ethics Committee Approval: This study was approved by the Kartal Dr. Lütfi Kırdar City Hospital Clinical Research Ethics Committee (Date: 27.07.2022, Decision No: 2022/514/230/18).
Peer-review: Externally peer-reviewed.
Authorship Contributions: Concept: R.A., M.K.; Design: R.A., M.K.; Supervision: R.A., M.K.; Resource: H.A., U.D.H.; Materials: H.A., U.D.H.; Data: R.A., M.K., H.A., U.D.H.; Analysis: R.A., M.K., H.A., U.D.H.; Literature search: R.A., M.K.; Writing: R.A., M.K.; Critical revision: R.A., M.K., H.A., U.D.H.
Conflict of Interest: None declared.
Financial Disclosure: The authors declared that this study has received no financial support.
REFERENCES
- 1.Nable JV, Graham AC. Gastrointestinal bleeding. Emerg Med Clin North Am. 2016;34:309–25. doi: 10.1016/j.emc.2015.12.001. [DOI] [PubMed] [Google Scholar]
- 2.Tielleman T, Bujanda D, Cryer B. Epidemiology and risk factors for upper gastrointestinal bleeding. Gastrointest Endosc Clin N Am. 2015;25:415–28. doi: 10.1016/j.giec.2015.02.010. [DOI] [PubMed] [Google Scholar]
- 3.Laine L, Yang H, Chang SC, Datto C. Trends for incidence of hospitalization and death due to GI complications in the United States from 2001 to 2009. Am J Gastroenterol. 2012;107:1190–95. doi: 10.1038/ajg.2012.168. quiz 1196. [DOI] [PubMed] [Google Scholar]
- 4.Laine L, Barkun AN, Saltzman JR, Martel M, Leontiadis GI. ACG clinical guideline:Upper gastrointestinal and ulcer bleeding. Am J Gastroenterol. 2021;116:899–917. doi: 10.14309/ajg.0000000000001245. [DOI] [PubMed] [Google Scholar]
- 5.Ohmiya N. Management of obscure gastrointestinal bleeding:Comparison of guidelines between Japan and other countries. Dig Endosc. 2020;32:204–18. doi: 10.1111/den.13554. [DOI] [PubMed] [Google Scholar]
- 6.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. doi: 10.1111/den.12639. [DOI] [PubMed] [Google Scholar]
- 7.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. doi: 10.1016/j.gie.2011.06.024. [DOI] [PubMed] [Google Scholar]
- 8.Rockall TA, Logan RF, Devlin HB, Northfield TC. Risk assessment after acute upper gastrointestinal haemorrhage. Gut. 1996;38:316–21. doi: 10.1136/gut.38.3.316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ak R, Hökenek NM. Comparison of AIMS65 and glasgow blatchford scores in predicting mortality in patients with upper gastrointestinal bleeding. Rev Assoc Méd Bras (1992) 2021;67:766–70. doi: 10.1590/1806-9282.20210580. [DOI] [PubMed] [Google Scholar]
- 10.Chang A, Ouejiaraphant C, Akarapatima K, Rattanasupa A, Prachayakul V. Prospective comparison of the AIMS65 score, glasgow-blatchford score, and rockall score for predicting clinical outcomes in patients with variceal and nonvariceal upper gastrointestinal bleeding. Clin Endosc. 2021;54:211–21. doi: 10.5946/ce.2020.068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Abougergi MS, Charpentier JP, Bethea E, Rupawala A, Kheder J, Nompleggi D, et al. A prospective, multicenter study of the AIMS65 score compared with the glasgow-blatchford score in predicting upper gastrointestinal hemorrhage outcomes. J Clin Gastroenterol. 2016;50:464–9. doi: 10.1097/MCG.0000000000000395. [DOI] [PubMed] [Google Scholar]
- 12.Hyett BH, Abougergi MS, Charpentier JP, Kumar NL, Brozovic S, Claggett BL, et al. The AIMS65 score compared with the glasgow-blatchford score in predicting outcomes in upper GI bleeding. Gastrointest Endosc. 2013;77:551–7. doi: 10.1016/j.gie.2012.11.022. [DOI] [PubMed] [Google Scholar]
- 13.Kim MS, Choi J, Shin WC. AIMS65 scoring system is comparable to glasgow-blatchford score or rockall score for prediction of clinical outcomes for non-variceal upper gastrointestinal bleeding. BMC Gastroenterol. 2019;19:136. doi: 10.1186/s12876-019-1051-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.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. doi: 10.1016/j.gie.2015.10.021. [DOI] [PubMed] [Google Scholar]
- 15.Liu S, Zhang X, Walline JH, Yu X, Zhu H. Comparing the performance of the ABC, AIMS65, GBS, and pRS scores in predicting 90-day mortality or rebleeding among emergency department patients with acute upper gastrointestinal bleeding:A prospective multicenter study. J Transl Int Med. 2021;9:114–22. doi: 10.2478/jtim-2021-0026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Yaka E, Yilmaz S, Doğan NO, Pekdemir M. Comparison of the glasgow-blatchford and AIMS65 scoring systems for risk stratification in upper gastrointestinal bleeding in the emergency department. Acad Emerg Med. 2015;22:22–30. doi: 10.1111/acem.12554. [DOI] [PubMed] [Google Scholar]
- 17.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. doi: 10.1136/bmj.i6432. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Budimir I, Stojsavljević S, Baršić N, Bišćanin A, Mirošević G, Bohnec S, et al. Scoring systems for peptic ulcer bleeding:Which one to use?World J Gastroenterol. 2017;23:7450–8. doi: 10.3748/wjg.v23.i41.7450. [DOI] [PMC free article] [PubMed] [Google Scholar]