Defining the quality characteristics of endoscopy for acute variceal hemorrhage in cirrhosis


  • Potential conflict of interest: Nothing to report.

Cheung J, Soo I, Bastiampillai R, Zhu Q, Ma M. Urgent vs. non-urgent endoscopy in stable acute variceal bleeding. Am J Gastroenterol 2009;104:1125-1129. (Reprinted with permission.)


Objectives: The optimal timing of endoscopy with acute variceal bleeding (AVB) is unknown. The aim of this study was to evaluate the association between the timing of endoscopy and outcomes of stable AVB patients. Methods: Patients admitted at two tertiary-care centers with hemodynamically stable AVB from 1997 to 2006 were evaluated retrospectively. The primary outcome was mortality. Other recorded outcomes included stigmata at endoscopy, hemostasis, blood transfusions, rebleeding, renal function, hospitalization length, infection, transjugular intrahepatic portosystemic shunt use, and balloon tamponade use. Logistic regression analysis was used to assess the association of time to endoscopy with mortality. Outcome comparisons were also performed for three different urgency times (< or = vs. > 4 h, < or = vs. > 8 h, and < or = vs. > 12 h). Results: There were 210 patients with stable AVB, accounting for 52% of the total number of AVB patients. The mean (+/− s.d.) age was 55 (+/− 12) years. The mean presenting systolic blood pressure and heart rate were 121 (+/− 16) mm Hg and 98 (+/− 20) bpm, respectively. Esophageal varices accounted for 91% (n = 191) of variceal bleeding. The mean time to endoscopy was 12 (+/− 12) h. The overall hemostasis rate after endoscopy was 97% (n = 203). The mortality rate was 9.5% (n = 20). There was no significant association of time to endoscopy with mortality (odds ratio, OR, 1.0; 95% confidence interval, CI, 0.92-1.08; P = 0.91). Significant independent predictors for mortality were lower albumin (OR, 0.82; 95% CI, 0.73-0.93; P = 0.001), infection during admission (OR, 8.9; 95% CI, 2.5-31.6; P < 0.001), and higher model end-stage liver disease (MELD) (OR, 1.17; 95% CI, 1.06-1.29; P = 0.002). There was no difference in outcomes with different urgency times. Conclusions: For patients who present with hemodynamically stable variceal bleeding, hemostasis after endoscopy is high, and the time to endoscopy does not appear to be associated with mortality.


Acute variceal hemorrhage (AVH) is a potentially lethal complication of portal hypertension affecting patients with cirrhosis; mortality estimates range from 10% to 30% per bleeding episode. Clinical practice guidelines have defined processes of care for the management of AVH, including the use of systemic vasoconstrictor agents, therapeutic endoscopy, and broad-spectrum antibiotics.1 However, there is evidence to date showing that compliance with treatment guidelines for AVH can be improved.2-5 Furthermore, the ability to identify the timeliness of care for AVH has not been examined until recently.

In this retrospective study, Cheung and colleagues6 sought to identify whether the timing of endoscopy was associated with mortality in hemodynamically stable patients with AVH. In their cohort of 210 patients, endoscopy was performed within 4 hours in 57 patients (27%), within 8 hours in 102 patients (48%), and within 12 hours in 134 patients (64%). Successful hemostasis from band ligation occurred in 97% of cases with an overall rebleeding rate of 19%. The in-hospital overall mortality rate after AVH was 9.5%, with multiorgan failure from sepsis being the leading cause rather than liver failure or recurrent AVH. Ultimately, multivariable analysis failed to show any significant relationship between the time to endoscopy and mortality.

The importance of timing for lifesaving interventions has been well documented for other diseases; for instance, a door-to-balloon time < 90 minutes for primary percutaneous coronary intervention (PCI) is associated with increased survival for patients suffering from acute myocardial infarction. Now, almost 90% of patients undergo PCI within 90 minutes of their arrival at the hospital across the United States.7 Factors including access, availability, and procedure volume determine, in part, the effectiveness of PCI. Similar arguments could be made for the management of AVH.

Availability of Endoscopy.

Several investigations have examined the potential impact of weekday availability of endoscopy versus weekend availability on clinical outcomes for AVH. Notably, there do not appear to be important differences in the rates of in-hospital mortality with respect to the day of admission and the availability of endoscopic services.8 Whether an association exists between weekend admission and increased mortality due to variceal bleeding over longer term follow-up (e.g., at 30 days or 1 year) warrants further investigation.

Variation in the Timing of Endoscopy.

There are documented variations in the timing of emergent endoscopy, and this appears to reflect the severity of a patient's presentation at first glance. According to the study by Myers et al.,8 the odds of mortality from AVH decreased by 6% with each additional day that endoscopy was delayed. Conversely, endoscopy on the day of admission was associated with a 45% increase in the odds of death. These findings suggest that clinical judgment results in appropriate triaging when sick individuals receive accelerated endoscopy, whereas endoscopy is delayed in those who are less severely ill. The absence of clinical and laboratory details within administrative data sets, however, limits our ability to judge whether hemodynamic parameters and conventional disease severity scores such as the Child-Turcotte-Pugh and Model for End-Stage Liver Disease scores on admission influence the timing of endoscopy.

Recently, there has been interest in quantifying the optimal duration between the initial presentation and therapeutic endoscopy (the door-to-scope time) for AVH. Although consensus expert opinion recommends endoscopy within 12 hours of presentation,1 clinicians may also opt to use pharmacological therapy before they perform endoscopy.9 The current study by Cheung et al.6 identified a mean time of 12 hours, yet no association with mortality was observed. Notably, the time to endoscopy in their cohort varied widely (from 1 to 78 hours), and patients with hemodynamic instability were excluded; this probably neutralized the effect of the time to endoscopy on outcome. Conversely, a mean time to endoscopy ≤ 15 hours was significantly associated with improved survival among 312 patients in an independent population.10 However, the door-to-scope time was not a highly sensitive (72%) or specific (59%) indicator of mortality because the Model for End-Stage Liver Disease score on admission, the failure to control bleeding during initial esophagogastroduodenoscopy, and the presence of hematemesis were more influential in determining mortality. Additional studies are required to ensure that rapid endoscopy is being performed for all patients with evidence of severe AVH.

Impact of the Case Volume on AVH Outcomes.

The relationship between the quality and the case volume has been studied extensively with the general notion that more experience could reduce population mortality and improve the efficiency of care. In contrast to other acute conditions, a significant relationship has not been identified between the volume and the outcomes after AVH.11, 12 Issues of inadequate risk adjustment and the absence of key predictors within claims data have likely contributed to the negative findings.

Variation in the Overall Process of Care for AVH.

From the standpoint of endoscopy, there appears to be broad consensus on the use of variceal band ligation versus sclerotherapy in the treatment of AVH.8 However, the use of antibiotic prophylaxis and systemic vasoconstrictors is more variable for AVH.2-6 Surprisingly, this degree of variation in the process of care has not been associated with increased mortality from AVH. The case mix and the severity of disease likely play significant roles, and their influence on outcomes also deserves further study.