Is timely endoscopy the answer for cost-effective management of acute upper gastrointestinal bleeding?

Endoscopy plays a major role in the management of acute upper gastrointestinal bleeding (UGIB) as it is currently accepted to be an effective procedure for the reduction of the rate of rebleeding, the need for surgery, and mortality [1]. However, because the safety of a very early endoscopy has been questioned and due to the associated organizational needs and related costs, diverse attempts have been made over the years to determine the best combination of peri- and intra-endoscopy procedures (“how”?), “when” to provide endoscopy, and the patients “to whom” endoscopy must be offered to achieve a cost-effective model of healthcare. It seems, however, that there is still room for improvement. In fact, even the group involved in the most recent recommendations for the management of acute nonvariceal UGIB [2] were only able to include the statements “endoscopy should be recommended within 24 hours of presentation for most patients” and “prognostic scales to stratify patients according to their death and rebleeding risks” should be used.

In 2001, a systematic review [3] found only three randomized trials in which early endoscopy was the intervention under study but the studies were considered to be of low methodological quality, with heterogeneous inclusion criteria and definitions of early endoscopy (from prompt to within 24 hours). The authors of the review concluded, based on observational and prospective uncontrolled trials, that in high risk patients (a definition that also varied across studies) rebleeding and surgical rates were improved when early endoscopy was provided but no change to mortality was observed. They recommended well-designed trials to be conducted and since then at least two randomized trials [4] [5] have compared endoscopy undertaken at less than 12 hours with endoscopy at 12 – 24 hours with no significant reduction in rebleeding, surgical or mortality rates being observed. It seems that any effort to clarify the best time to perform endoscopy is complicated by the variations in routine clinical practice and the type of patients treated. Probably the best evidence will come from studies with very well defined and standard criteria and procedures. Should we move from “the sooner the better” as concluded in 2001 to the concept of “a timely endoscopy” and do we have the studies to answer these questions?

In this issue of Endoscopy, Jairath et al. [6] presented the results of a prospective audit in 212 hospitals in the UK. The study addressed whether the time to upper gastrointestinal endoscopy is related to rebleeding, need for surgery, length of hospital stay, and mortality in 4478 patients with acute nonvariceal UGIB. Even though no effect was observed in the mortality or surgery rate, earlier endoscopy did increase the efficiency of care: a substantial reduction in risk-adjusted length of hospital stay and potentially improved control of hemorrhage in higher risk patients were observed, which may support the routine use of early endoscopy with potential savings. This supports previous reports that in all risk groups early endoscopy significantly reduces the length of hospital stay [7]. It was also noted that early endoscopy (sooner than 24 hours) was effective, but in high-risk patients (those who will need endoscopic therapy), an even earlier endoscopy seems to be beneficial in terms of rebleeding. Again, what seems more rational is to argue in favor of an adjustment of the concept of early endoscopy according to patient risk for rebleeding and other outcomes – the concept of “a timely endoscopy.”

These kinds of studies with large numbers of patients, if valid, will clarify the inconclusive results from previous single-centered, small or nonrandomized trials. Moreover, they will facilitate discussion of potential drawbacks of studies and of clinical practice reporting, perhaps providing the opportunity for additional studies and further improvement! In this case, 50 % of the patients underwent endoscopy after 24 hours of bleeding, no clear criteria for blood transfusion were reported, and there appeared to be a potential misuse of resources with 5 % of low-risk lesions (too many?) and only 61 % of high-risk lesions (not enough) being subjected to endoscopic treatment; this perhaps suggests variation in the definitions of risk predictors. Again, current recommendations [1] include diverse suggestions concerning pharmacological and endoscopic methods but no statement relating to the wider management of care in these patients, such as access to healthcare, minimal training, and skill set of emergency teams. As Spiegel concluded in 2001 [2], we also need standards in this setting.

In another contribution to this issue of Endoscopy, de Groot et al. [8] present a very important paper on the second part of this treatment concept – timely endoscopy. The authors summarized 16 studies, published from 1987 to 2009, that described diverse scores, all of which considered pre-endoscopic variables and five of which also included endoscopic features, all with the aim of predicting different clinical outcomes after UGIB. Importantly, substantial heterogeneity and the absence of external validation and impact analysis for these scores were noticeable. Different variables were considered, with co-morbidity (11 studies or 17 if liver disease and obesity are included), blood pressure (10 studies), heart rate (5 studies), and age (5 studies) being the most frequently considered among pre-endoscopic parameters. The authors were able to suggest specific scores for different outcomes based on good or high methodological quality, predictive power, and external validation. A score that would work for all outcomes, even if different cut-offs were included, would be welcomed. Adequate derivation and validation in different settings, using “time to endoscopy” as outcome, and all other clinical and organizational outcomes would also need to be considered.

We may therefore conclude that if endoscopy is provided in a timely manner, improvement in all major clinical outcomes and probably savings in the cost of care of patients with acute UGIB could be achieved. Nevertheless, we need more studies in this setting, particularly those that consider the complete process of care, as it is clear that the answer will vary according to the diversity of healthcare settings, including the proportion of high-risk patients coming to emergency care, health providers’ awareness of standards of practice, and the associated costs.

• References

• 1 Cook DJ, Guyatt GH, Salena BJ et al. Endoscopic therapy for acute nonvariceal upper gastrointestinal hemorrhage: a meta-analysis. Gastroenterology 1992; 102: 139-148
• 2 Barkun AN, Bardou M, Kuipers EJ et al. for the International Consensus Upper Gastrointestinal Bleeding Conference Group. International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding. Ann Intern Med 2010; 152: 101-113
• 3 Spiegel BM, Vakil NB, Ofman JJ. Endoscopy for acute nonvariceal upper gastrointestinal tract hemorrhage: is sooner better? A systematic review.. Arch Intern Med 2001; 161: 1393-1404
• 4 Bjorkman DJ, Zaman A, Fennerty MB et al. Urgent vs elective endoscopy for acute non-variceal upper-GI bleeding: an effectiveness study. Gastrointest Endosc 2004; 60: 1-8
• 5 Lin HJ, Wang K, Perng CL et al. Early or delayed endoscopy for patients with peptic ulcer bleeding. A prospective randomized study. J Clin Gastroenterol 1996; 22: 267-271
• 6 Jairath V, Kahan BC, Logan RFA et al. Outcomes following acute nonvariceal upper gastrointestinal bleeding in relation to time to endoscopy: results from a nationwide study. Endoscopy 2012; 44: 723-730
• 7 Lim LG, Ho KY, Chan YH et al. Urgent endoscopy is associated with lower mortality in high-risk but not low-risk nonvariceal upper gastrointestinal bleeding. Endoscopy 2011; 43: 300-306
• 8 de Groot NL, Bosman JH, Siersema PD et al. Prediction scores in gastrointestinal bleeding: a systematic review and quantitative appraisal. Endoscopy 2012; 44: 731-739