The Validation of Rockall Scoring System in Predicting Outcomes from Variceal Bleeding in Sudanese Patients: A Cross-sectional Study

Widad A Sharfeldeen¹, Ahmed Rafei^2*, Hinda H Mohammed³, Ola Emad Ahmed¹, Mohammad S Katta¹ and Abdelmoneim Eltayeib Abdo^4
*Correspondence: Ahmed Rafei, Department of Research, National Centre for Gastrointestinal and Liver Diseases, Khartoum, Sudan, Tel: + 00249905486627, Email:

Keywords

Rockall score • Upper GI bleeding • Variceal bleeding • Sudan

Introduction

Upper Gastrointestinal Bleeding (UGIB) is a prevalent issue, occurring at a rate of approximately 80 to 150 cases per 100,000 individuals annually. Its associated mortality rates are estimated to range from 2% to 15%. UGIB is defined as any form of blood loss originating from the gastrointestinal tract above the ligament of Treitz. Clinical manifestations may include hematemesis (vomiting bright red blood or coffee-ground-like material), hematochezia, or melena. Patients may also exhibit symptoms resulting from blood loss, such as fainting episodes, fatigue, and general weakness. UGIB can present as acute, occult, or obscure bleeding [1].

Among the potential causes of Upper Gastrointestinal Bleeding (UGIB), Peptic Ulcer Disease (PUD) is responsible for 40% to 50% of the cases. Other significant contributors include erosive esophagitis (11%), duodenitis (10%), varices (ranging from 5% to 30% depending on whether the population studied has a chronic liver disease), Mallory-Weiss tear (5% to 15%), and vascular malformations (5%) [2].

ariceal bleeding arises from enlarged veins (varices) situated at the intersection of the portal and systemic venous systems, typically occurring in the lower part of the esophagus or the upper part of the stomach. Individuals experiencing variceal hemorrhage commonly exhibit evident Upper Gastrointestinal Bleeding (UGIB), marked by symptoms such as hematemesis and/or melena. The prognosis for patients suffering from variceal bleeding is closely tied to the intensity of the bleeding incident, liver function, and portal venous pressure [3]. Mortality from variceal bleeding has been found to be as high as 11 per 100 people [4]. In the emergency department, acute variceal hemorrhage is a major cause of mortality and can be challenging to treat. The mortality in a bleeding episode depends on the severity of the underlying liver disease, with rates being less than 10% in well-compensated cirrhotic patients and over 70% in those with advanced cirrhosis. The risk of re-bleeding is also significant, reaching up to 80% within a year [5]. The mortality rate following the first bleeding episode was found to be between 17% and 57% [6], and inadequate management can lead to a high risk of re-bleeding [7].

The Rockall system has demonstrated its efficacy and reliability in predicting the likelihood of re-bleeding and mortality. The Rockall scores incorporate various factors, including the patient's age, the presence of shock as evaluated through systolic blood pressure readings and pulse rate, the existence and severity of coexisting medical conditions, as well as the presence of visible signs of haemorrhage [8]. Previous investigations and validations of the scoring system have revealed that individuals with a score of ≤2 are linked with a notably low incidence of bleeding, recurrence, and mortality, thus enabling their potential management as outpatients. Patients with a Rockall score of ≥3 have a high risk of re-bleeding and mortality, and require hospitalization and intensive care. This offers clinicians an opportunity for a more targeted approach to patient care, tailored to their assessed risk of complications subsequent to the initial Upper Gastrointestinal Bleeding (UGIB) [9].

The Rockall scoring system is particularly useful in predicting the outcomes of variceal bleeding, which is a major cause of UGIB in Sudan [10]. In Sudan, the majority of cases of variceal bleeding are due to schistosomial peri-portal fibrosis [10]. Which is caused by the development of T cell-dependent granuloma around schistosome mansoni eggs, leading to hepatic peri-portal fibrosis [11]?

This study aims to evaluate the validation of the Rockall scoring system in predicting the outcomes of variceal bleeding among Sudanese patients presenting to the Mohammed Salih Idris Bleeding Center (MSIBC) which is the national tertiary referral bleeding center. The results of this study could help to improve the management of patients with variceal bleeding in Sudan, and reduce the risk of re-bleeding and mortality.

The work has been reported in line with the STROCSS criteria [12].

Methods

This cross-sectional, observational study was conducted at the Mohammed Salih Idris Bleeding Center (MSIBC) in Khartoum, Sudan. A total of 150 patients were enrolled in the study. The study comprised patients aged 18 years and older, presenting with upper Gastrointestinal (GI) bleeding of variceal origin, who visited the outpatient and referral clinic at the center and reported primary complaints of upper GI bleeding (UGIB), including hematemesis and/or melena. Patients with upper GI bleeding resulting from causes other than varices, as well as those who declined to participate in the study, were excluded

Variables

The data were collected through interviews conducted by trained medical doctors using structured questionnaire forms, and it was categorized into three sections. The first part included demographic data such as age, gender, occupation, and residence. The second part encompassed the patient's medical history, comprising a history of Schistosomal infection, Hepatitis B and Hepatitis C infections, the cause of portal hypertension, past instances of variceal bleeding, history of alcohol consumption, presenting complaints, history of past endoscopic interventions, and the number of sessions conducted.

The findings from the upper GI endoscopy were collected and included the grade of esophageal varices, endoscopic diagnosis, and evidence of bleeding. The third part encompassed the international Rockall score (Table 1) [13], which included factors such as age, evidence of shock, other co-morbid conditions, the cause of bleeding, and evidence of recent bleeding. Other significant variables collected during the study included the interventions conducted, complications that developed thereafter, occurrences of bleeding recurrence, duration of hospital stay, and instances of mortality.

The upper GI endoscopy was performed within 24 hours of presentation by an experienced gastrointestinal physician or gastrointestinal surgeon. The location and grade of varices were recorded, along with evidence of recent bleeding and details of the intervention method. Following the administration of appropriate pharmacological and endoscopic therapy, patients were monitored for evidence of early re-bleeding or mortality until their discharge from the hospital.

Data analysis

The data were characterized by their range, mean along with standard deviation (±SD), median, frequencies denoting the number of cases, and relative frequencies presented as percentages where appropriate. A paired t-test was utilized to compare quantitative variables for parametric data, while the Chi-square (χ²) test was employed for comparing categorical data. In cases where the expected frequency was below 5, the exact test was used. A probability value (p-value) below 0.05 was considered to be statistically significant. All statistical computations were conducted using Statistical Package for the Social Sciences (SPSS) version 26.0 for Microsoft Windows (IBM SPSS Statistics, Armonk, NY, USA).

Ethics

All participants provided written, informed consent. Strict confidentiality protocols were implemented to safeguard the privacy and anonymity of the participants. The study received approval from the Mohammed Salih Idris Bleeding Center ethical committee, Khartoum, Sudan.

Results

Seventy-nine percent (n=119) of the patients were aged between 18 and 60 years old, with a male predominance (117, 78%) reported. Additionally, the majority of the patients resided in central states (134, 89%) (Table 2).

Concerning presenting complaints, 70 patients (46.7%) reported hematemesis, 22 patients (14.6%) reported melena, and 58 patients (38.7%) reported experiencing both symptoms. Peri-portal fibrosis was identified as the most common cause (101, 67.3%), followed by cirrhosis (24, 16%), with both causes coexisting in 23 patients (15.3%). Two patients (1.3%) had other causes. A positive history of schistosomiasis was noted in 90 patients (60%), while sixteen patients (10.7%) had a positive hepatitis B screening, and none tested positive for hepatitis C. Fifty-eight percent (n=87) of the patients reported alcohol consumption.

In terms of the grade of esophageal varices, 4 patients (2.7%) had grade I esophageal varices, 19 patients (12.7%) had grade II, 67 patients (44.7%) had grade III, 47 patients (31.3%) had grade IV, and 13 patients (8.7%) were diagnosed with fundal varices.

Fifty-four percent (n=81) of the patients experienced upper gastrointestinal (UGI) bleeding for the first time without a history of prior UGI bleeding. Concerning the past history of endoscopic interventions, 94 patients (62.7%) had not undergone any intervention, 53 patients (35.3%) had received sclerotherapy, and only three patients (2%) had undergone band ligation (Table 2).

The mean Rockall score was calculated to be 3.83±1.99. Regarding the individual element scores, 66 patients (44%) exhibited a pulse rate exceeding 100, while 33 patients (22%) had a systolic blood pressure below 100. Co-morbid conditions were identified in 10 patients (6.7%) with coexistent liver failure, 3 patients (2%) with renal failure, and 2 patients (1.3%) with both conditions. Additionally, three patients tested positive for metastatic cancer, and one patient had ischemic heart disease (IHD). The majority of the patients (129, 86%) had no co-morbidities.

Out of the patients observed, 42 (28%) exhibited no evidence of bleeding, while 91 patients (60.7%) had visible blood, 15 patients (10%) had an adherent clot, and spurting vessels were found in one patient. As for complications, two patients (1.3%) experienced encephalopathy, one patient (1.3%) had renal failure, and the remaining patients (97.3%) did not develop any complications.

There was no statistically significant association between Rockall score and neither the grade of esophageal varices (p value is 0.7) nor with past medical history of endoscopic intervention (p value is 0.538) nor with the current intervention (p value 0.951) (Table 3). In contrast, there was statistically significant association between Rockall score and each of the following: number of cessions (p<0.014), complications (p<0.0001), bleeding recurrence (p<0.001), duration of hospital stay (p<0.002), death (p<0.0001) (Table 3).

Discussion

The majority of individuals within the study population were below the age of 60, a critical demographic representing an important socioeconomic and productive age group. A prior Sudanese study yielded similar findings regarding age [14]. The male gender predominated in the study sample, attributed to the prevalence of males in the agricultural sector in Sudan, which poses a risk of contracting Schistosomal infection. More than half of the study population had a history of Schistosomal infection, with two-thirds developing peri-portal fibrosis as a consequence of this past exposure. Consequently, the major cause of portal hypertension in Sudan is Schistosomal peri-portal fibrosis.

The mean Rockall score was calculated to be 3.83±1.99, which is lower than the result described in another study where the mean Rockall score ranged between 4 and 5 [15]. The incidence of bleeding, as evidenced by endoscopy, was notably higher compared to a previous study conducted in Canada [16], suggesting the severity of variceal bleeding in Sudan.

Regarding the pre-endoscopic Rockall score, our study indicated a 0% mortality rate in individuals with a pre-endoscopic Rockall score of 0 or 1. There was also a statistically significant association between mortality and scores of 2 and above. These results align with a prospective study conducted by Phang TSM, et al., which reported a mortality rate of 3.2% in the low-risk group [17]. The risk of bleeding recurrence was observed in all patients with scores ranging from 0 to 6, with an overall estimated risk higher compared to a previous study conducted by Sarwar S, et al. [18].

Seven patients, representing 4.7% of the study population succumbed to mortality (Table 4). The mortality rate in our study was lower than that reported in a study conducted in Lahore, where the mortality rate was 6.7% [19]. This difference is likely related to the underlying cause of portal hypertension, as cirrhotic patients commonly decompensate following variceal bleeding, contrasting with the outcomes observed in patients with peri-portal fibrosis [20].

Regarding the pre-endoscopic Rockall score, this study demonstrated a 0% mortality rate in patients with a score of 0 or 1. A statistically significant association between mortality and scores of 2 and above was also observed (p<0.001). These findings are in line with a prospective study conducted by Phang TSM, et al., which validated the pre-endoscopic Rockall score and reported a mortality rate of less than 1% in patients with a score of 0 or 1 [17]. The study further revealed a correlation between increasing initial Rockall scores and mortality, suggesting the potential for patient triage into different care models based on their scores.

Another prospective study, involving 358 patients, evaluated the effectiveness of the initial Rockall risk scoring system in predicting re-bleeding and mortality among patients with esophageal varices or peptic ulcers [20]. The study highlighted zero mortality in patients presenting with acute Upper Gastrointestinal Bleeding (UGIB) due to peptic ulcers or varices, and with an initial pre-endoscopic score of 0 to 1. It also confirmed a significant correlation between hospital mortality and scores of 2 and above, which aligns with the findings of our study.

This study, however, encountered some limitations, such as a relatively small sample size, and the lack of post-discharge follow-up to monitor delayed complications or mortality should be noted. Nonetheless, efforts were made to minimize recall bias by utilizing a widely used, valid, and reliable instrument that provided a concise and thorough assessment of upper gastrointestinal bleeding. To overcome this limitation and promote a more universal applicability of research findings, future investigations could benefit from adopting larger sample sizes.

The study represents a pioneering effort in the Sudanese context, addressing a significant research gap. Before this investigation, there was a notable absence of studies validating the Rockall scoring system specifically in the context of variceal bleeding in Sudan. This lack of localized validation studies impeded the establishment of a reliable foundation for clinical decision-making in managing variceal bleeding cases. Recognizing this gap, our study took the initiative to lay the groundwork for future research endeavors in the region. It stands as a significant contribution, providing a vital cornerstone for subsequent research in Sudanese populations, thereby enhancing the understanding and management of variceal bleeding in Sudan.

Conclusion

The findings underscore the importance of this scoring system in predicting outcomes and guiding clinical decision-making for Sudanese patients presenting with variceal bleeding. The study's identification of peri-portal fibrosis as a predominant cause of portal hypertension in Sudan contributes essential knowledge to the local landscape. The observed association between Rockall scores and various clinical outcomes, including mortality and bleeding recurrence, provides a valuable basis for risk stratification and tailored patient management. Importantly, the study addresses a critical research gap by being the first of its kind in Sudan, thereby laying a foundation for future investigations and serving as a cornerstone for enhancing the understanding and management of variceal bleeding in this specific population.

Acknowledgement

Patient consent

Written informed consent was obtained from the patient for publication and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request.

Ethical approval

The study received approval from the Mohammed Salih Idris Bleeding Center ethical committee, Khartoum, Sudan on 2 January 2012.

Conflict of Interest

None.

Assistance with the Study

None.

Financial Support and Sponsorship

None.

References

1. Kamboj, Amrit K., Patrick Hoversten and Cadman L. Leggett. "Upper gastrointestinal bleeding: Etiologies and management." Mayo Clin Proc 94 (2019): 697-703.

Google Scholar, Crossref, Indexed at

2. https://www.ncbi.nlm.nih.gov/books/NBK470300/
3. Thabut, Dominique and Brigitte Bernard-Chabert. "Management of acute bleeding from portal hypertension." Best Pract Res Clin Gastroenterol 21 (2007): 19-29.

Google Scholar, Crossref, Indexed at

4. Child, Charles Gardner. "Surgery and portal hypertension." Major Probl Clin Surg (1964): 1-85.

Google Scholar, Indexed at

5. Zaman, Atif. "Current management of esophageal varices." Curr Treat Options Gastroentero 6 (2003): 499-507.

Google Scholar, Crossref, Indexed at

6. Habib, Adil and Arun J. Sanyal. "Acute variceal hemorrhage." Gastrointest Endosc Clin N Am 17 (2007): 223-252.

Google Scholar, Crossref, Indexed at

7. Rockall, T. A., R. F. Logan, H. B. Devlin and TC8675081 Northfield. "Risk assessment after acute upper gastrointestinal haemorrhage." Gut 38 (1996): 316-321.

Google Scholar, Crossref, Indexed at

8. Rockall, T. A., R. F. A. Logan, H. B. Devlin and T. C. Northfield. "Incidence of and mortality from acute upper gastrointestinal haemorrhage in the United Kingdom." Bmj 311 (1995): 222-226.

Google Scholar, Crossref, Indexed at

9. Bozkurt, Mehmet Abdussamet, Kıvanç Derya Peker, Mustafa Gökhan Unsal and Hakan Yırgın, et al. "The Importance of Rockall scoring system for upper gastrointestinal bleeding in long-term follow-up." Indian J Surg 79 (2017): 188-191.

Google Scholar, Crossref, Indexed at

10. Gasim, B., S. S. Fedail, A. M. Musaad and S. M. Salih, et al. "Endoscopic sclerotherapy for bleeding oesophageal varices: experience in Sudan." Trop Gastroenterol 23 (2002): 107-109.

Google Scholar, Indexed at

11. Da Silva, L. C., Pedro Paulo Chieffi and Flair José Carrilho. "Schistosomiasis mansoni–clinical features." Gastroenterol Hepatol 28 (2005): 30-39.

Google Scholar, Crossref, Indexed at

12. Mathew, Ginimol and Riaz Agha. "STROCSS 2021: strengthening the reporting of cohort, cross-sectional and case-control studies in surgery." IJS Short Reports 6 (2021): e35.

Google Scholar, Crossref, Indexed at

13. Vreeburg, E. M., C. B. Terwee, P. Snel and E. A. J. Rauws, et al. "Validation of the Rockall risk scoring system in upper gastrointestinal bleeding." Gut 44 (1999): 331-335.

Google Scholar, Crossref, Indexed at

14. Sanders, David S., Mike J. Perry, Simon GW Jones and Elaine McFarlane, et al. "Effectiveness of an upper-gastrointestinal haemorrhage unit: a prospective analysis of 900 consecutive cases using the Rockall score as a method of risk standardisation Eur J Gastroenterol Hepatol 16 (2004): 487-494.

Google Scholar, Crossref, Indexed at

15. Das, Ananya, Tamir Ben–Menachem, Farees T. Farooq and Gregory S. Cooper, et al. "Artificial neural network as a predictive instrument in patients with acute nonvariceal upper gastrointestinal hemorrhage." Gastroenterology 134 (2008): 65-74.

Google Scholar, Crossref, Indexed at

16. Salih, H., M. A. M. Ibnouf, A. Siddig and A. Masaad. "Rockall score of the acute upper gastrointestinal bleeding patients the experience in Sudan." Sudan J Med Sci 4 (2009).

Google Scholar, Crossref, Indexed at

17. Phang, Tony SM, Vadim Vornik and Richard Stubbs. "Risk assessment in upper gastrointestinal haemorrhage: Implications for resource utilisation." N Z Med J 113 (2000): 331.

Google Scholar, Indexed at

18. Sarwar, Shahid, Akif Dilshad, Anwaar Ahmad Khan and Altaf Alam, et al. "Predictive value of Rockall score for rebleeding and mortality in patients with variceal bleeding." J Coll Physicians Surg Pak 7 (2007): 253-256.

Google Scholar, Indexed at

19. East, James E., David Westaby and Amit Rastogi. "Management of varices in cirrhosis."  N Engl J Med 362 (2010): 2331-2332.

Google Scholar, Crossref, Indexed at

20. Sanders, D. S., M. J. Carter, R. J. Goodchap and S. S. Cross, et al. "Prospective validation of the Rockall risk scoring system for upper GI hemorrhage in subgroups of patients with varices and peptic ulcers." Am J Gastroenterol 97 (2002): 630-635.

Google Scholar, Crossref, Indexed at