Original Article
Validity of the Rockall scoring system after endoscopic therapy for bleeding peptic ulcer: a prospective cohort study

https://doi.org/10.1016/j.gie.2005.06.042Get rights and content

Background

The Rockall scoring system was developed in unselected patients, the majority of whom did not receive endoscopic therapy. The aim of this study was to assess the validity of the Rockall system in high-risk patients who undergo endoscopic therapy for peptic ulcer hemorrhage.

Methods

Rockall scores were calculated in 247 patients with major peptic ulcer bleeding entered into a randomized trial of endoscopic therapy. The observed rates of recurrent bleeding and mortality after endoscopic therapy were compared with predicted rates derived from Rockall's study group. The validity of the Rockall system was assessed in terms of calibration and discrimination.

Results

Rates of recurrent bleeding and mortality after endoscopic therapy increased with an increasing Rockall score. Observed rates of recurrent bleeding and mortality were below predicted rates, and calibration of the Rockall system was poor (Mantel-Haenszel chi square = 25.8, p < 0.0001 for recurrent bleeding; Mantel-Haenszel chi square = 15.1, p < 0.0001 for death). For the prediction of recurrent bleeding, the area under the receiver operating characteristic curve was low (63.4%), but the system was satisfactory when predicting mortality (area under the resulting curve, 84.3%).

Conclusions

After endoscopic therapy for a bleeding peptic ulcer, the Rockall scoring system can identify patients at high risk of death, but it is inadequate for the prediction of recurrent bleeding.

Section snippets

Patients and methods

Between November 1996 and January 2001, 247 patients with a bleeding peptic ulcer were recruited to a randomized, controlled trial that compared endoscopic therapy with a heat probe plus human thrombin injection and endoscopic therapy with a heat probe plus placebo injection. The results of the study have been published.11 All patients presented with significant upper-GI hemorrhage and at least one of the following clinical risk factors: age over 60 years, Hb less than 10 g/dL, the presence of

Results

The Rockall scores for all 247 patients in the trial followed a fairly symmetrical distribution (Fig. 1).

The mean score for the entire sample was 5.8. Because all patients had peptic ulcer disease with active bleeding or stigmata of recent hemorrhage, the minimum Rockall score was 3. Thirty-six patients (15%) had recurrent bleeding within 30 days of endoscopic therapy, and 22 (9%) died. The outcome in relation to Rockall score is shown in Table 2. There was a significant relationship between

Discussion

The Rockall scoring system was developed from an analysis of data obtained from a large audit of patients who presented with upper-GI bleeding because of a wide range of causes, and only a minority of patients received endoscopic therapy. The current analysis was designed to determine whether the Rockall system is applicable to the subset of patients with high-risk ulcer bleeding who require endoscopic hemostatic therapy. It also was desirable to identify a cutoff point below which patients

References (16)

There are more references available in the full text version of this article.

Cited by (65)

  • The association between blood urea nitrogen to creatinine ratio and mortality in patients with upper gastrointestinal bleeding

    2018, Arab Journal of Gastroenterology
    Citation Excerpt :

    Therapeutic decisions such as disposition of patients with UGIB, urgent endoscopic examinations or blood transfusions were different depending on the prognosis and severity of the UGIB. With the recognition of pre-endoscopic risk factors, several scoring systems such as AIMS65, Glasgow-Blatchford Risk Score (GBS), and Rockall score have been designed to predict the mortality of UGIB patients [4,5,31,32]. Within the Glasgow-Blatchford risk score, BUN is used as one of the parameters and accounts for 6 points of the total GBS score.

  • Efficacy of Over-the-Scope Clips in Management of High-Risk Gastrointestinal Bleeding

    2018, Clinical Gastroenterology and Hepatology
    Citation Excerpt :

    Most OTSCs were placed earlier in the hospital course (mean LOS before OTSC, 3.9 d; SD, ±8.0 d; mean LOS after OTSC, 7.4 d; SD, ±6.8 d). Among upper GI bleeds, the mean Rockall score was 6.9 (SD, ±1.4; >7 carries high-risk mortality),15 and the modified Blatchford score was 10.1 (SD, ±2.5; >9 carries high-risk re-intervention),14 showing the heightened clinical severity of our OTSC patient population. Overall, the mean hemoglobin nadir before OTSC placement on index admission was 7.5 (SD, ±1.8), and the resuscitation required was aggressive because the mean total number of packed red blood cell transfusions required was 4.9 U (SD, ±5.8 U) primarily before OTSC placement (mean, 4.4 U; SD, ±5.7 U) as compared with after placement (mean, 0.5 U; SD, ±0.9 U).

  • C-reactive protein as a prognostic indicator for rebleeding in patients with nonvariceal upper gastrointestinal bleeding

    2015, Digestive and Liver Disease
    Citation Excerpt :

    There are some prognostic scales to predict rebleeding and mortality, such as the Blatchford and Rockall scores [3,4]. These scoring systems have been validated in many studies, but have better discriminative ability for mortality than rebleeding [21–25], which requires additional predictive markers. Since CRP was first discovered in the serum of patients infected with Streptococcus pneumonia [26], it has been recognized clinically as a nonspecific inflammatory marker.

View all citing articles on Scopus

This study and N. I. Church's salary were supported by a grant from the Scottish National Blood Transfusion Service.

See CME section; p. 676.

View full text