- Email: [email protected]
Second look endoscopy in acute non-variceal upper gastrointestinal bleeding
Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
Contents lists available at ScienceDirect
Best Practice & Research Clinical Gastroenterology
7
Second look endoscopy in acute non-variceal upper gastrointestinal bleeding Philip Wai Yan Chiu, MD, FRCSEd, Professor * Institute of Digestive Disease, Department of Surgery, The Chinese University of Hong Kong, Prince of Wales Hospital, 30-32 Ngan Shing Street, Shatin, N.T., Hong Kong SAR, China
a b s t r a c t Keywords: Acute non-variceal upper gastrointestinal bleeding Second look endoscopy Cost-effectiveness
Bleeding peptic ulcer remained an important cause of hospitalization worldwide. Primary endoscopic hemostasis achieved more than 90% of initial hemostasis for bleeding peptic ulcer. Recurrent bleeding amounted to 15% after therapeutic endoscopy, and rebleeding is an important risk factor to peptic ulcer related mortality. Routine second look endoscopy was one of the strategies targeted at prevention of rebleeding. The objective of second look endoscopy was to treat persistent stigmata of recent hemorrhage before rebleeding. Three meta-analyses showed that performance of routine second look endoscopy significantly reduced ulcer rebleeding especially when the endoscopic therapy was performed with thermal coagulation. Two cost-effectiveness analyses, however, demonstrated that selective instead of routine second look endoscopy is the most cost-effective approach to prevent ulcer rebleeding. While international consensus and guidelines did not recommend routine performance of second look endoscopy for prevention of ulcer rebleeding, further research should focus on identification of patients with high risk of rebleeding and investigate the effect of selective second look endoscopy in prevention of rebleeding among these patients. Ó 2013 Elsevier Ltd. All rights reserved.
* Tel.: þ852 26322627; fax: þ852 26377974. E-mail address: [email protected]. 1521-6918/$ – see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bpg.2013.09.009
906
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
Introduction Upper gastrointestinal (GI) bleeding remained a common and important cause of hospitalization worldwide [1]. Non-variceal upper gastrointestinal bleeding accounted for 250,000 to 300,000 of hospital admissions and 15,000 to 30,000 deaths per year [2]. In United Kingdom, the annual admission for upper gastrointestinal bleeding was 84.6 per 100,000 hospital admissions [3]. Bleeding peptic ulcer is the most common cause for non-variceal upper gastrointestinal bleeding. Despite the global decline in the incidence of Helicobacter pylori related peptic ulcer, the incidence of bleeding peptic ulcers remained enormous [4]. The incidence of peptic ulcer disease ranged from 0.10 to 0.19% basing on hospitalization data from two European studies [5,6]. With the increasing incidence of NSAID and aspirin related peptic ulcers, admissions for complicated peptic ulcers remained stable or slightly increased [7]. The mortality related to non-variceal upper GI bleeding remained around 6%, despite the advancement in the endoscopic treatment and management [8]. While majority of the mortality were related to non-bleeding causes, 18.4% of the mortality was bleeding related. Most of these bleeding related mortalities were due to uncontrolled bleeding or rebleeding. A risk prediction score on mortality basing on 3000 patients with bleeding peptic ulcers showed that ulcer rebleeding and need of surgery were important risk factor with an odds ratio of 1.63 [9]. A review on risk prediction in acute upper GI bleeding also showed that rebleeding is an important cause of mortality [10]. Zimmerman et al analysed the prognostic factors to acute upper GI bleeding and found that age older than 75 years, blood in stomach, high serum creatinine level and persistent or recurrent bleeding were predictors to mortality [11]. In the management of acute upper GI bleeding, primary endoscopic hemostasis and prevention of rebleeding became important focus to prevent mortality. Endoscopic therapy for bleeding peptic ulcers Primary endoscopic therapy had replaced surgery as the initial treatment for upper gastrointestinal bleeding. Two meta-analysis in the 1990s confirmed that endoscopic therapy is effective in achieving primary hemostasis for bleeding peptic ulcers [12,13]. The common types of endoscopic treatment for bleeding peptic ulcers included injection, thermal and mechanical therapies. Endoscopic injection of epinephrine was one of the earliest treatments for bleeding peptic ulcers [14]. Though effective as a primary treatment, endoscopic injection of epinephrine was associated with 30% risk of rebleeding. A recent meta-analysis comparing clip against injection and thermal therapy for treatment of non-variceal upper GI bleeding included 1156 patients from 15 studies [15]. Definitive hemostasis was significantly higher with endoscopic hemoclip alone or endoscopic hemoclip and injection as compared to injection alone. This also led to a reduction in the need of surgery. There was no difference in the rate of definitive hemostasis, rebleeding and need of surgery between thermo-coagulation and hemoclip. Marmo et al conducted a meta-analysis to compare dual therapy against monotherapy in endoscopic treatment of high risk bleeding ulcers [16]. This meta-analysis included 2472 patient s from 20 prospective randomized trials which compared mono against dual therapy for ulcer hemostasis. Compared to control, a combination of endoscopic therapies significantly reduced risk of rebleeding and need of surgery, with a trend towards reduction in risk of death. From these meta-analyses, primary endoscopic therapy should be the recommended treatment for acute non-variceal upper GI bleeding. Meanwhile, endoscopic therapy should be standardized as either thermo-coagulation or hemoclip alone, or a combination of injection plus thermo-coagulation or hemoclip. Despite the high success of initial hemostasis using endoscopic therapy, 10–15% of patients were at risk of developing rebleeding. Prevention of ulcer rebleeding after successful primary endoscopic hemostasis became an important issue. Clinical evidence on second look endoscopy to prevent peptic ulcer rebleeding There are numerous strategies being investigated to prevent peptic ulcer rebleeding. A better hemostatic method during primary endoscopic therapy should improve the rate of permanent hemostasis. Once primary endoscopic hemostasis was secured, two strategies were investigated extensively in prevention of further bleeding, including performance of scheduled second endoscopy and administration of potent acid suppression therapy.
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
907
Scheduled second endoscopy surveys the bleeding ulcer at 16–24 h after initial hemostasis. The aim of this surveillance is to pick up persistent stigmata of recent hemorrhage and commence another round of treatment to prevent rebleeding [17]. Controversies abound upon the effectiveness of second look endoscopy in preventing peptic ulcer rebleeding [18–23]. Villanueva et al reported that patients receiving second look endoscopy had 7% absolute risk reduction in rebleeding when compared to control(no routine scheduled second look endoscopy), though this did not reach statistical significance (95% CI 1.3– 11.1) [18]. Messmann et al conducted a multi-center prospective randomized trial to investigate programmed endoscopic follow-up with retreatment for bleeding peptic ulcers, and 105 patients were recruited [19]. Endoscopic retreatment would be commenced to ulcers showing persistent endoscopic stigmata of the Forrest I, IIa, and IIb. The authors found that there was no difference in rebleeding, length of hospital stay, and number of units of blood transfused between those who received daily follow-up endoscopy and retreatment against those who had clinical observation alone. Saeed et al adopted a different approach where selective second look endoscopy was performed for patients at high risk of rebleeding [20]. The selection was based on Baylor bleeding score, which categorized patients into high or low risk of rebleeding basing on age, number and severity of comorbidities, site of bleeding and stigmata of recent hemorrhage. They then randomized patients categorized as high risk of rebleeding to either a routine second endoscopic retreatment or conservative management. There was a significant difference in rebleeding between the groups that received routine second look endoscopic retreatment (0%) against those who had clinical observation (24%). Lin et al conducted another study on non-selective programmed second endoscopy [21] One hundred and thirty-one patients were randomly assigned to receive a second look endoscopy or conservative management. The authors found that there is a significant reduction in recurrent bleeding in the group of patients that received a second look endoscopy. Our group conducted a prospective randomized trial investigating the effect of scheduled second endoscopy for prevention of ulcer rebleeding in 194 patients [22]. The performance of second endoscopy with appropriate retreatment at 16–24 h significantly reduced the risk of rebleeding from 13.8% to 5%. Two meta-analyses were conducted to examine the role of second look endoscopy in prevention of peptic ulcer rebleeding, which included the prospective randomized trials described above [24,25] (Table 1). Marmo et al performed a systematic review of randomized controlled trials investigating the role of second look endoscopy for bleeding peptic ulcers between 1990 and 2000. A total of 4 prospective randomized trials were included, three of which endoscopic injection of epinephrine or fibrin glue was used while in another study heater probe was employed for endoscopic treatment of bleeding peptic ulcer. Second look endoscopy with retreatment significant reduced the risk of ulcer rebleeding with an odds ratio of 0.64 (95% CI 0.44–0.95). The number needed to treat for second look endoscopy was 16. All the studies did not use adjunctive proton pump inhibitors except one, which did not show difference in ulcer rebleeding between second endoscopy and control group [24]. Chiu et al included 5 prospective randomized trials in a meta-analysis on scheduled second endoscopy for prevention of peptic ulcer rebleeding [25]. Although there was significant heterogeneity between these Table 1 Summary of the results of meta-analysis comparing second look endoscopy against conservative management for peptic ulcer bleeding. Meta-analyses
Marmo et al [23] Chiu et al [24] Tsoi et al [25] Tsoi et al [25] a a
No of trials
No of patients Second look endoscopy
No second look
Rebleeding Second look endoscopy
No second look
Need of surgery Second look endoscopy
No second look
Second look endoscopy
No second look
4
393
392
12%
18%
16 (4%)
22 (5.7%)
17 (4.3%)
21 (5.2%)
5
283
274
10.9%
19.7%
1
6
–
–
5
541
533
119
115
110/533 (20.6%) 18/115 (15.6%)
16/374 (4.3%)
2
90/514 (16.6%) 5/119 (4.2%)
23/371 (6.2%) –
16/374 (4.3%) –
19/371 (5.1%) –
–
Mortality
Results of second look endoscopy compared to conservative treatment with thermal coagulation for endoscopic hemostasis.
908
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
trials as shown from analysis using fix and random effect models, second look endoscopy significantly reduce risk of rebleeding. The number needed to scope in order to prevent ulcer rebleeding, however, ranged from 4 to 11. The performance of second endoscopy is not without risk, as further endoscopic treatment could result in higher chance of complications such as perforation, especially when thermal coagulation was repeatedly applied [26,27]. By repeating second endoscopy, most patients just had another survey over a non-bleeding ulcer without therapy. This will induce additional discomfort to patients as well as workload to endoscopists. Furthermore, most randomized trials on second look endoscopy were limited by small sample size and the employment of single modality of endoscopic therapy. Tsoi et al conducted a meta-analysis with a view to compare the effect of second look endoscopy using thermal or injection therapy [28]. From 5 prospective randomized trials, 119 patients received routine second look endoscopy with thermal coagulation, 374 with endoscopic injection and 505 patients had single endoscopic therapy alone. These five randomized trials were published from 1991 to 2003, with the sample sizes ranged from 40 to 536 patients. Thermal therapy with second look endoscopy significantly reduced rebleeding when compared to single endoscopic treatment alone (4.2% vs 15.7%), while second look endoscopy with injection alone showed no difference in ulcer rebleeding compared to single endoscopic treatment (17.6% vs 20.8%) (Table 1). It is clear that routine second look endoscopy should be performed with thermal coagulation and not injection alone. This reflected the limitation of the meta-analyses of routine second look endoscopy as most of the randomized trials reported the use of single modality endoscopic therapy. Recent meta-analysis recommended the use of combination therapy instead of single therapy for primary endoscopic hemostasis in the management of bleeding peptic ulcers [16], as combination endoscopic therapy significantly reduced the rate of ulcer rebleeding. Cost-effectiveness of second look endoscopy From our previous randomized trial, only 36% of patients who received second look endoscopy required endoscopic retreatment [22]. The performance of second look endoscopy would certainly impose significant patient discomfort, and the role of second endoscopy to survey for persistent endoscopic stigmata without accompanying endoscopic treatment is doubtful. Moreover, second look endoscopy might induce significant workload to endoscopists as well as the endoscopy unit. Spiegel et al conducted a cost-effectiveness analysis comparing various strategies to minimize recurrent peptic ulcer hemorrhage [29]. Four strategies were compared through decision analysis, including: 1. Follow patients clinically after hemostasis and repeat endoscopy on for patients with evidence of rebleeding; 2. Administer intravenous PPI after hemostasis and repeat endoscopy only in patients with clinical signs of rebleeding; 3. Perform second look endoscopy at 24 h in all patients with successful endoscopic hemostasis; and 4. Perform selective second look endoscopy at 24 h only in patients at high risk of rebleeding. The cost estimation was based on Medicare reimbursement, while the effectiveness was measured by the rate of rebleeding, need of surgery or prevention of death. It was found the selective second look endoscopy was the most effective and least expensive strategy with 91% effectiveness with number needed to treat (NNT) of 10, a reduction in the cost incurred as compared to the standard treatment of endoscopy for suspected rebleeding. However, the cost-effective analysis on selective second look endoscopy was based one single prospective randomized trial which only recruited 40 patients. Although the rebleeding rate for high risk patients was significantly reduced after selective second endoscopy with retreatment from 24% to 0%, the outcomes of those categorized as low risk were not reported. Selective second look endoscopy for high risk patients remained controversial, and further research should be conducted to investigate the clinical efficacy of this approach. Imperiale et al estimated resource implications of second look endoscopy for bleeding peptic ulcer through a decision model and cost-effectiveness analysis using a decision model based on literature probabilities [30]. Overall, routine second look endoscopy reduced ulcer rebleeding from 16% to 10% with no effect on other clinical outcomes including need of surgery and hospital mortality. Under the Medicare reimbursement system, the cost to prevent 1 case of rebleeding with routine second look endoscopy was nearly US$ 13,000. With the threshold analysis, it was found that population would need a rebleeding rate of 31% to offset the cost difference between routine second look and no second look endoscopy. The authors concluded that routine second look endoscopy is not recommended after
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
909
therapeutic endoscopy for bleeding peptic ulcer unless the rebleeding rate was 31% or greater. In the modern era of management of bleeding peptic ulcer, with the application of combination therapy for ulcer hemostasis and use of high dose PPI infusion, the rate of rebleeding should range between 5 and 10% only [25,26,31]. However, selective second look endoscopy may be the cost-effective approach as the study from Saeed et al showed that this selective approach to perform second endoscopy categorized as high risk using Baylor score significantly reduced rebleeding from 24% to 0% [20]. In an international consensus statement for management of bleeding peptic ulcer, routine second look endoscopy was not recommended not recommended unless it was performed in selected high risk patients or when initial endoscopic examination was incomplete due to technical reasons, as a result of insufficient evidence to definitively prove its benefit and due to the existence of PPI as an alternative mechanism to reduce risk of rebleeding [31]. From the Asia Pacific consensus on management of bleeding peptic ulcers, second look endoscopy was again not recommended as a routine practice but should be reserved for selected high risk patients [32]. The Asia Pacific panel believes that there is room for further research to identify specific group of patients whom second look endoscopy with thermal or mechanical therapy may be beneficial. From a prospective randomized trial of 240 patients, Lau et al showed that adjunctive high dose PPI infusion significantly reduced peptic ulcer rebleeding [33]. Adjunctive high dose PPI infusion became the preferred strategy to prevent ulcer rebleeding as patients would not need to receive another surveillance endoscopy. Although the performance of routine second look endoscopy achieved a similar rate of peptic ulcer rebleeding compared to high dose PPI infusion through indirect comparison across studies, there was no direct comparative study on the cost-effectiveness between the two approaches. Conclusions Clinical studies and meta-analysis showed that routine second look endoscopy with thermal coagulation therapy significantly reduced peptic ulcer rebleeding. However, since most studies compared second look endoscopy without PPI, routine second look endoscopy is not recommended as standard clinical practice as it is not likely to be cost-effective when compared to management with proton pump inhibitors. Future research should concentrate on identification of patients who are at high risk of recurrent bleeding despite adequate PPI and investigate the effect of selective second look endoscopy among these patients.
Practice points Routine second look endoscopy with thermal coagulation reduced peptic ulcer rebleeding. Routine second look endoscopy was not recommended as standard clinical practice for prevention of peptic ulcer rebleeding as it imposed significant workload to endoscopists and the need of endoscopic retreatment was low. Cost-effective analysis showed that the cost for routine second look endoscopy shall be offset only if the rebleeding rate was greater than 30%. In the modern era of therapeutic endoscopy for management of bleeding peptic ulcer, the rebleeding rate was usually ranged from 5 to 10%. Selective second look endoscopy in patients categorized as high risk of rebleeding significant reduced rebleeding from 24% to 0%.
Research agenda Future research should be conducted to investigate the effectiveness of selective second look endoscopy in preventing peptic ulcer rebleeding among patients with high risk of rebleeding. Large scale studies should focus on establishing and validating various risk scoring system in prediction for peptic ulcer rebleeding.
910
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
Disclosure The author has no conflict of interest for disclosure.
References *[1] Rockall TA, Logan RFA, Devlin HB, Northfield TC. Influence the practice and outcome in acute upper gastrointestinal haemorrhage. Gut 1997;41:606–11. [2] Gilbert DA. Epidemiology of upper gastrointestinal bleeding. Gastrointest Endosc 1990;36(Suppl.):S8–13. [3] Crooks CJ, West J, Card TR. Upper gastrointestinal haemorrhage and deprivation: a nationwide cohort study of health inequality in hospital admissions. Gut 2012;61(4):514–20. [4] Sung JJY, Kuipers EJ, El-Serag HB. Systematic review: the global incidence and prevalence of peptic ulcer disease. Aliment Pharmacol Ther 2009;29:938–46. [5] Hein HO, Suadicani P, Gyntelberg F. Genetic markers for peptic ulcer. A study of 3387 men aged 54 to 74 years: the Copenhagen Male Study. Scand J Gastroenterol 1997;32:16–21. [6] Kang JY, Elders A, Majeed A, Maxwell JD, Bardhan KD. Recent trends in hospital admissions and mortality rates for peptic ulcer in Scotland 1982–2002. Aliment Pharmacol Ther 2006;24:65–79. [7] Paimela H, Paimela L, Myllykangas-Luosujarvi R, Kivilaakso E. Current features of peptic ulcer disease in Finland: incidence of surgery, hospital admissions and mortality for the disease during the past twenty-five years. Scand J Gastroenterol 2002;37:399–403. *[8] Sung JJ, Tsoi KK, Ma TK, Yung MY, Lau JY, Chiu PW. Causes of mortality in patients with peptic ulcer bleeding: a prospective cohort study of 10,428 cases. Am J Gastroenterol 2010 Jan;105(1):84–9. *[9] Chiu PW, Ng EK, Cheung FK, Chan FK, Leung WK, Wu JC, et al. Predicting mortality in patients with bleeding peptic ulcers after therapeutic endoscopy. Clin Gastroenterol Hepatol 2009 Mar;7(3):311–6. [10] Chiu PW, Ng EK. Predicting poor outcome from acute upper gastrointestinal hemorrhage. Gastroenterol Clin North Am 2009 Jun;38(2):215–30. [11] Zimmerman J, Siguencia J, Tsvang E, et al. Predictors to mortality in patients admitted to hospital for acute upper gastrointestinal hemorrhage. Scand J Gastroenterol 1995;30:327–31. [12] Cook DJ, Guyatt GH, Salena BJ, et al. Endoscopic therapy for acute non-variceal upper gastrointestinal hemorrhage: a meta-analysis. Gastroenterology 1992;102:139–48. [13] Sacks HS, Chalmers TC, Blum AL, et al. Endoscopic hemostasis: an effective therapy for bleeding peptic ulcer. JAMA 1990; 264(4):494–9. [14] Chung SC, Leung JW, Steele RJ, Crofts TJ, Li AK. Endoscopic injection of adrenaline for actively bleeding ulcers: a randomised trial. Br Med J (Clin Res Ed) 1988 Jun 11;296(6637):1631–3. *[15] Sung JJ, Tsoi KK, Lai LH, Wu JC, Lau JY. Endoscopic clipping versus injection and thermo-coagulation in the treatment of non-variceal upper gastrointestinal bleeding: a meta-analysis. Gut 2007;56:1364–72. *[16] Marmo R, Rotondano G, Piscopo R, Bianco MA, D’Angella R, Cipoletta L. Dual therapy versus monotherapy in the endoscopic treatment of high risk bleeding ulcers: a meta-analysis of controlled trials. Am J Gastroenterol 2007; 102(2):279–89. [17] Saeed ZA. Second thoughts about second-look endoscopy for ulcer bleeding? Endoscopy 1998;30:650–2. [18] Villanueva C, Balanzo J, Torras X, et al. Value of second-look endoscopy after injection therapy for bleeding peptic ulcer: a prospective and randomized trial. Gastrointest Endosc 1994. [19] Messmann H, Schaller P, Andus T, et al. Effect of programmed endoscopic follow-up examinations on the rebleeding rate of gastric or duodenal ulcers treated by injection therapy: a prospective, randomized controlled trial. Endoscopy 1998;30: 583–9. 40:34–9. *[20] Saeed ZA, Cole RA, Ramirez FC, et al. Endoscopic retreatment after successful initial hemostasis prevents ulcer rebleeding: a prospective randomized trial. Endoscopy 1996;28:288–94. [21] Lin CK, Lai KH, Lo GH, et al. The value of second-look endoscopy after endoscopic injection therapy for bleeding peptic ulcer. Gastroenterology 1996;110:A177. *[22] Chiu PWY, Lam CYW, Lee SW, Kwong KH, Lam SH, Lee DTY, et al. Effect of scheduled second therapeutic endoscopy on peptic ulcer rebleeding: a prospective randomized trial. Gut 2003;52:1403–7. *[23] Marmo R, Rotondano G, Bianco MA, Pscopo R, Prisco A, Cipolletta L. Outcome of endoscopic treatment for peptic ulcer bleeding: Is a second look necessary? A meta-analysis. Gastrointest Endosc 2003;57:62–7. [24] Chiu PW, Lau TS, Kwong KH, Suen DT, Kwok SP. Impact of programmed second endoscopy with appropriate retreatment on peptic ulcer rebleeding: a systematic review. Ann Coll Surgeons Hong Kong 2003;7(4):106–15. [25] Tsoi KK, Chiu PW, Sung JJ. Endoscopy for upper gastrointestinal bleeding: is routine second-look necessary? Nat Rev Gastroenterol Hepatol 2009 Dec;6(12):717–22. [26] Chiu PW, Sung JJ. High risk ulcer bleeding: when is second-look endoscopy recommended? Clin Gastroenterol Hepatol 2010 Aug;8(8):651–4. [27] Chung SCS. Preventing ulcer rebleeding: the role of second-look endoscopy. Can J Gastroenterol 1999;13(5):409–11. [28] Tsoi KK, Chan HC, Chiu PW, Pau CY, Lau JY, Sung JJ. Second look endoscopi with thermal coagulation or injection for peptic ulcer bleeding: a meta-analysis. J Gastroenterol Hepatol 2010 Jan;25(1):8–13. *[29] Spiegel BM, Ofman JJ, Woods K, Vakil NB. Minimizing recurrent peptic ulcer hemorrhage after endoscopic hemostasis: the cost-effectiveness of competing strategies. Am J Gastroenterol 2003;98(1):86–97. *[30] Imperiale TF, Kong N. Second-look endoscopy for bleeding peptic ulcer disease: a decision-effectiveness and costeffectiveness analysis. J Clin Gastroenterol 2012;46(9):e71–5. [31] Barkun A, Bardou M, Kulpers E, Sng JJ, Hunt RH, Martel M, et al. International consensus recommendations on the management of patients with non-variceal upper gastrointestinal bleeding. Ann Intern Med 2010;152:101–13.
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
911
[32] Sung JJ, Chan FK, Chen M, Ching JY, Ho KY, Kachinorn U, et al. Asia Pacific working group consensus on non-variceal upper gastrointestinal bleeding. Gut 2011;60:1170–7. [33] Lau JY, Sung JJ, Lee KK, Yung MY, Wong SK, Wu JC, et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers. N Engl J Med 2000 Aug 3;343(5):310–6.
Contents lists available at ScienceDirect
Best Practice & Research Clinical Gastroenterology
7
Second look endoscopy in acute non-variceal upper gastrointestinal bleeding Philip Wai Yan Chiu, MD, FRCSEd, Professor * Institute of Digestive Disease, Department of Surgery, The Chinese University of Hong Kong, Prince of Wales Hospital, 30-32 Ngan Shing Street, Shatin, N.T., Hong Kong SAR, China
a b s t r a c t Keywords: Acute non-variceal upper gastrointestinal bleeding Second look endoscopy Cost-effectiveness
Bleeding peptic ulcer remained an important cause of hospitalization worldwide. Primary endoscopic hemostasis achieved more than 90% of initial hemostasis for bleeding peptic ulcer. Recurrent bleeding amounted to 15% after therapeutic endoscopy, and rebleeding is an important risk factor to peptic ulcer related mortality. Routine second look endoscopy was one of the strategies targeted at prevention of rebleeding. The objective of second look endoscopy was to treat persistent stigmata of recent hemorrhage before rebleeding. Three meta-analyses showed that performance of routine second look endoscopy significantly reduced ulcer rebleeding especially when the endoscopic therapy was performed with thermal coagulation. Two cost-effectiveness analyses, however, demonstrated that selective instead of routine second look endoscopy is the most cost-effective approach to prevent ulcer rebleeding. While international consensus and guidelines did not recommend routine performance of second look endoscopy for prevention of ulcer rebleeding, further research should focus on identification of patients with high risk of rebleeding and investigate the effect of selective second look endoscopy in prevention of rebleeding among these patients. Ó 2013 Elsevier Ltd. All rights reserved.
* Tel.: þ852 26322627; fax: þ852 26377974. E-mail address: [email protected]. 1521-6918/$ – see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bpg.2013.09.009
906
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
Introduction Upper gastrointestinal (GI) bleeding remained a common and important cause of hospitalization worldwide [1]. Non-variceal upper gastrointestinal bleeding accounted for 250,000 to 300,000 of hospital admissions and 15,000 to 30,000 deaths per year [2]. In United Kingdom, the annual admission for upper gastrointestinal bleeding was 84.6 per 100,000 hospital admissions [3]. Bleeding peptic ulcer is the most common cause for non-variceal upper gastrointestinal bleeding. Despite the global decline in the incidence of Helicobacter pylori related peptic ulcer, the incidence of bleeding peptic ulcers remained enormous [4]. The incidence of peptic ulcer disease ranged from 0.10 to 0.19% basing on hospitalization data from two European studies [5,6]. With the increasing incidence of NSAID and aspirin related peptic ulcers, admissions for complicated peptic ulcers remained stable or slightly increased [7]. The mortality related to non-variceal upper GI bleeding remained around 6%, despite the advancement in the endoscopic treatment and management [8]. While majority of the mortality were related to non-bleeding causes, 18.4% of the mortality was bleeding related. Most of these bleeding related mortalities were due to uncontrolled bleeding or rebleeding. A risk prediction score on mortality basing on 3000 patients with bleeding peptic ulcers showed that ulcer rebleeding and need of surgery were important risk factor with an odds ratio of 1.63 [9]. A review on risk prediction in acute upper GI bleeding also showed that rebleeding is an important cause of mortality [10]. Zimmerman et al analysed the prognostic factors to acute upper GI bleeding and found that age older than 75 years, blood in stomach, high serum creatinine level and persistent or recurrent bleeding were predictors to mortality [11]. In the management of acute upper GI bleeding, primary endoscopic hemostasis and prevention of rebleeding became important focus to prevent mortality. Endoscopic therapy for bleeding peptic ulcers Primary endoscopic therapy had replaced surgery as the initial treatment for upper gastrointestinal bleeding. Two meta-analysis in the 1990s confirmed that endoscopic therapy is effective in achieving primary hemostasis for bleeding peptic ulcers [12,13]. The common types of endoscopic treatment for bleeding peptic ulcers included injection, thermal and mechanical therapies. Endoscopic injection of epinephrine was one of the earliest treatments for bleeding peptic ulcers [14]. Though effective as a primary treatment, endoscopic injection of epinephrine was associated with 30% risk of rebleeding. A recent meta-analysis comparing clip against injection and thermal therapy for treatment of non-variceal upper GI bleeding included 1156 patients from 15 studies [15]. Definitive hemostasis was significantly higher with endoscopic hemoclip alone or endoscopic hemoclip and injection as compared to injection alone. This also led to a reduction in the need of surgery. There was no difference in the rate of definitive hemostasis, rebleeding and need of surgery between thermo-coagulation and hemoclip. Marmo et al conducted a meta-analysis to compare dual therapy against monotherapy in endoscopic treatment of high risk bleeding ulcers [16]. This meta-analysis included 2472 patient s from 20 prospective randomized trials which compared mono against dual therapy for ulcer hemostasis. Compared to control, a combination of endoscopic therapies significantly reduced risk of rebleeding and need of surgery, with a trend towards reduction in risk of death. From these meta-analyses, primary endoscopic therapy should be the recommended treatment for acute non-variceal upper GI bleeding. Meanwhile, endoscopic therapy should be standardized as either thermo-coagulation or hemoclip alone, or a combination of injection plus thermo-coagulation or hemoclip. Despite the high success of initial hemostasis using endoscopic therapy, 10–15% of patients were at risk of developing rebleeding. Prevention of ulcer rebleeding after successful primary endoscopic hemostasis became an important issue. Clinical evidence on second look endoscopy to prevent peptic ulcer rebleeding There are numerous strategies being investigated to prevent peptic ulcer rebleeding. A better hemostatic method during primary endoscopic therapy should improve the rate of permanent hemostasis. Once primary endoscopic hemostasis was secured, two strategies were investigated extensively in prevention of further bleeding, including performance of scheduled second endoscopy and administration of potent acid suppression therapy.
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
907
Scheduled second endoscopy surveys the bleeding ulcer at 16–24 h after initial hemostasis. The aim of this surveillance is to pick up persistent stigmata of recent hemorrhage and commence another round of treatment to prevent rebleeding [17]. Controversies abound upon the effectiveness of second look endoscopy in preventing peptic ulcer rebleeding [18–23]. Villanueva et al reported that patients receiving second look endoscopy had 7% absolute risk reduction in rebleeding when compared to control(no routine scheduled second look endoscopy), though this did not reach statistical significance (95% CI 1.3– 11.1) [18]. Messmann et al conducted a multi-center prospective randomized trial to investigate programmed endoscopic follow-up with retreatment for bleeding peptic ulcers, and 105 patients were recruited [19]. Endoscopic retreatment would be commenced to ulcers showing persistent endoscopic stigmata of the Forrest I, IIa, and IIb. The authors found that there was no difference in rebleeding, length of hospital stay, and number of units of blood transfused between those who received daily follow-up endoscopy and retreatment against those who had clinical observation alone. Saeed et al adopted a different approach where selective second look endoscopy was performed for patients at high risk of rebleeding [20]. The selection was based on Baylor bleeding score, which categorized patients into high or low risk of rebleeding basing on age, number and severity of comorbidities, site of bleeding and stigmata of recent hemorrhage. They then randomized patients categorized as high risk of rebleeding to either a routine second endoscopic retreatment or conservative management. There was a significant difference in rebleeding between the groups that received routine second look endoscopic retreatment (0%) against those who had clinical observation (24%). Lin et al conducted another study on non-selective programmed second endoscopy [21] One hundred and thirty-one patients were randomly assigned to receive a second look endoscopy or conservative management. The authors found that there is a significant reduction in recurrent bleeding in the group of patients that received a second look endoscopy. Our group conducted a prospective randomized trial investigating the effect of scheduled second endoscopy for prevention of ulcer rebleeding in 194 patients [22]. The performance of second endoscopy with appropriate retreatment at 16–24 h significantly reduced the risk of rebleeding from 13.8% to 5%. Two meta-analyses were conducted to examine the role of second look endoscopy in prevention of peptic ulcer rebleeding, which included the prospective randomized trials described above [24,25] (Table 1). Marmo et al performed a systematic review of randomized controlled trials investigating the role of second look endoscopy for bleeding peptic ulcers between 1990 and 2000. A total of 4 prospective randomized trials were included, three of which endoscopic injection of epinephrine or fibrin glue was used while in another study heater probe was employed for endoscopic treatment of bleeding peptic ulcer. Second look endoscopy with retreatment significant reduced the risk of ulcer rebleeding with an odds ratio of 0.64 (95% CI 0.44–0.95). The number needed to treat for second look endoscopy was 16. All the studies did not use adjunctive proton pump inhibitors except one, which did not show difference in ulcer rebleeding between second endoscopy and control group [24]. Chiu et al included 5 prospective randomized trials in a meta-analysis on scheduled second endoscopy for prevention of peptic ulcer rebleeding [25]. Although there was significant heterogeneity between these Table 1 Summary of the results of meta-analysis comparing second look endoscopy against conservative management for peptic ulcer bleeding. Meta-analyses
Marmo et al [23] Chiu et al [24] Tsoi et al [25] Tsoi et al [25] a a
No of trials
No of patients Second look endoscopy
No second look
Rebleeding Second look endoscopy
No second look
Need of surgery Second look endoscopy
No second look
Second look endoscopy
No second look
4
393
392
12%
18%
16 (4%)
22 (5.7%)
17 (4.3%)
21 (5.2%)
5
283
274
10.9%
19.7%
1
6
–
–
5
541
533
119
115
110/533 (20.6%) 18/115 (15.6%)
16/374 (4.3%)
2
90/514 (16.6%) 5/119 (4.2%)
23/371 (6.2%) –
16/374 (4.3%) –
19/371 (5.1%) –
–
Mortality
Results of second look endoscopy compared to conservative treatment with thermal coagulation for endoscopic hemostasis.
908
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
trials as shown from analysis using fix and random effect models, second look endoscopy significantly reduce risk of rebleeding. The number needed to scope in order to prevent ulcer rebleeding, however, ranged from 4 to 11. The performance of second endoscopy is not without risk, as further endoscopic treatment could result in higher chance of complications such as perforation, especially when thermal coagulation was repeatedly applied [26,27]. By repeating second endoscopy, most patients just had another survey over a non-bleeding ulcer without therapy. This will induce additional discomfort to patients as well as workload to endoscopists. Furthermore, most randomized trials on second look endoscopy were limited by small sample size and the employment of single modality of endoscopic therapy. Tsoi et al conducted a meta-analysis with a view to compare the effect of second look endoscopy using thermal or injection therapy [28]. From 5 prospective randomized trials, 119 patients received routine second look endoscopy with thermal coagulation, 374 with endoscopic injection and 505 patients had single endoscopic therapy alone. These five randomized trials were published from 1991 to 2003, with the sample sizes ranged from 40 to 536 patients. Thermal therapy with second look endoscopy significantly reduced rebleeding when compared to single endoscopic treatment alone (4.2% vs 15.7%), while second look endoscopy with injection alone showed no difference in ulcer rebleeding compared to single endoscopic treatment (17.6% vs 20.8%) (Table 1). It is clear that routine second look endoscopy should be performed with thermal coagulation and not injection alone. This reflected the limitation of the meta-analyses of routine second look endoscopy as most of the randomized trials reported the use of single modality endoscopic therapy. Recent meta-analysis recommended the use of combination therapy instead of single therapy for primary endoscopic hemostasis in the management of bleeding peptic ulcers [16], as combination endoscopic therapy significantly reduced the rate of ulcer rebleeding. Cost-effectiveness of second look endoscopy From our previous randomized trial, only 36% of patients who received second look endoscopy required endoscopic retreatment [22]. The performance of second look endoscopy would certainly impose significant patient discomfort, and the role of second endoscopy to survey for persistent endoscopic stigmata without accompanying endoscopic treatment is doubtful. Moreover, second look endoscopy might induce significant workload to endoscopists as well as the endoscopy unit. Spiegel et al conducted a cost-effectiveness analysis comparing various strategies to minimize recurrent peptic ulcer hemorrhage [29]. Four strategies were compared through decision analysis, including: 1. Follow patients clinically after hemostasis and repeat endoscopy on for patients with evidence of rebleeding; 2. Administer intravenous PPI after hemostasis and repeat endoscopy only in patients with clinical signs of rebleeding; 3. Perform second look endoscopy at 24 h in all patients with successful endoscopic hemostasis; and 4. Perform selective second look endoscopy at 24 h only in patients at high risk of rebleeding. The cost estimation was based on Medicare reimbursement, while the effectiveness was measured by the rate of rebleeding, need of surgery or prevention of death. It was found the selective second look endoscopy was the most effective and least expensive strategy with 91% effectiveness with number needed to treat (NNT) of 10, a reduction in the cost incurred as compared to the standard treatment of endoscopy for suspected rebleeding. However, the cost-effective analysis on selective second look endoscopy was based one single prospective randomized trial which only recruited 40 patients. Although the rebleeding rate for high risk patients was significantly reduced after selective second endoscopy with retreatment from 24% to 0%, the outcomes of those categorized as low risk were not reported. Selective second look endoscopy for high risk patients remained controversial, and further research should be conducted to investigate the clinical efficacy of this approach. Imperiale et al estimated resource implications of second look endoscopy for bleeding peptic ulcer through a decision model and cost-effectiveness analysis using a decision model based on literature probabilities [30]. Overall, routine second look endoscopy reduced ulcer rebleeding from 16% to 10% with no effect on other clinical outcomes including need of surgery and hospital mortality. Under the Medicare reimbursement system, the cost to prevent 1 case of rebleeding with routine second look endoscopy was nearly US$ 13,000. With the threshold analysis, it was found that population would need a rebleeding rate of 31% to offset the cost difference between routine second look and no second look endoscopy. The authors concluded that routine second look endoscopy is not recommended after
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
909
therapeutic endoscopy for bleeding peptic ulcer unless the rebleeding rate was 31% or greater. In the modern era of management of bleeding peptic ulcer, with the application of combination therapy for ulcer hemostasis and use of high dose PPI infusion, the rate of rebleeding should range between 5 and 10% only [25,26,31]. However, selective second look endoscopy may be the cost-effective approach as the study from Saeed et al showed that this selective approach to perform second endoscopy categorized as high risk using Baylor score significantly reduced rebleeding from 24% to 0% [20]. In an international consensus statement for management of bleeding peptic ulcer, routine second look endoscopy was not recommended not recommended unless it was performed in selected high risk patients or when initial endoscopic examination was incomplete due to technical reasons, as a result of insufficient evidence to definitively prove its benefit and due to the existence of PPI as an alternative mechanism to reduce risk of rebleeding [31]. From the Asia Pacific consensus on management of bleeding peptic ulcers, second look endoscopy was again not recommended as a routine practice but should be reserved for selected high risk patients [32]. The Asia Pacific panel believes that there is room for further research to identify specific group of patients whom second look endoscopy with thermal or mechanical therapy may be beneficial. From a prospective randomized trial of 240 patients, Lau et al showed that adjunctive high dose PPI infusion significantly reduced peptic ulcer rebleeding [33]. Adjunctive high dose PPI infusion became the preferred strategy to prevent ulcer rebleeding as patients would not need to receive another surveillance endoscopy. Although the performance of routine second look endoscopy achieved a similar rate of peptic ulcer rebleeding compared to high dose PPI infusion through indirect comparison across studies, there was no direct comparative study on the cost-effectiveness between the two approaches. Conclusions Clinical studies and meta-analysis showed that routine second look endoscopy with thermal coagulation therapy significantly reduced peptic ulcer rebleeding. However, since most studies compared second look endoscopy without PPI, routine second look endoscopy is not recommended as standard clinical practice as it is not likely to be cost-effective when compared to management with proton pump inhibitors. Future research should concentrate on identification of patients who are at high risk of recurrent bleeding despite adequate PPI and investigate the effect of selective second look endoscopy among these patients.
Practice points Routine second look endoscopy with thermal coagulation reduced peptic ulcer rebleeding. Routine second look endoscopy was not recommended as standard clinical practice for prevention of peptic ulcer rebleeding as it imposed significant workload to endoscopists and the need of endoscopic retreatment was low. Cost-effective analysis showed that the cost for routine second look endoscopy shall be offset only if the rebleeding rate was greater than 30%. In the modern era of therapeutic endoscopy for management of bleeding peptic ulcer, the rebleeding rate was usually ranged from 5 to 10%. Selective second look endoscopy in patients categorized as high risk of rebleeding significant reduced rebleeding from 24% to 0%.
Research agenda Future research should be conducted to investigate the effectiveness of selective second look endoscopy in preventing peptic ulcer rebleeding among patients with high risk of rebleeding. Large scale studies should focus on establishing and validating various risk scoring system in prediction for peptic ulcer rebleeding.
910
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
Disclosure The author has no conflict of interest for disclosure.
References *[1] Rockall TA, Logan RFA, Devlin HB, Northfield TC. Influence the practice and outcome in acute upper gastrointestinal haemorrhage. Gut 1997;41:606–11. [2] Gilbert DA. Epidemiology of upper gastrointestinal bleeding. Gastrointest Endosc 1990;36(Suppl.):S8–13. [3] Crooks CJ, West J, Card TR. Upper gastrointestinal haemorrhage and deprivation: a nationwide cohort study of health inequality in hospital admissions. Gut 2012;61(4):514–20. [4] Sung JJY, Kuipers EJ, El-Serag HB. Systematic review: the global incidence and prevalence of peptic ulcer disease. Aliment Pharmacol Ther 2009;29:938–46. [5] Hein HO, Suadicani P, Gyntelberg F. Genetic markers for peptic ulcer. A study of 3387 men aged 54 to 74 years: the Copenhagen Male Study. Scand J Gastroenterol 1997;32:16–21. [6] Kang JY, Elders A, Majeed A, Maxwell JD, Bardhan KD. Recent trends in hospital admissions and mortality rates for peptic ulcer in Scotland 1982–2002. Aliment Pharmacol Ther 2006;24:65–79. [7] Paimela H, Paimela L, Myllykangas-Luosujarvi R, Kivilaakso E. Current features of peptic ulcer disease in Finland: incidence of surgery, hospital admissions and mortality for the disease during the past twenty-five years. Scand J Gastroenterol 2002;37:399–403. *[8] Sung JJ, Tsoi KK, Ma TK, Yung MY, Lau JY, Chiu PW. Causes of mortality in patients with peptic ulcer bleeding: a prospective cohort study of 10,428 cases. Am J Gastroenterol 2010 Jan;105(1):84–9. *[9] Chiu PW, Ng EK, Cheung FK, Chan FK, Leung WK, Wu JC, et al. Predicting mortality in patients with bleeding peptic ulcers after therapeutic endoscopy. Clin Gastroenterol Hepatol 2009 Mar;7(3):311–6. [10] Chiu PW, Ng EK. Predicting poor outcome from acute upper gastrointestinal hemorrhage. Gastroenterol Clin North Am 2009 Jun;38(2):215–30. [11] Zimmerman J, Siguencia J, Tsvang E, et al. Predictors to mortality in patients admitted to hospital for acute upper gastrointestinal hemorrhage. Scand J Gastroenterol 1995;30:327–31. [12] Cook DJ, Guyatt GH, Salena BJ, et al. Endoscopic therapy for acute non-variceal upper gastrointestinal hemorrhage: a meta-analysis. Gastroenterology 1992;102:139–48. [13] Sacks HS, Chalmers TC, Blum AL, et al. Endoscopic hemostasis: an effective therapy for bleeding peptic ulcer. JAMA 1990; 264(4):494–9. [14] Chung SC, Leung JW, Steele RJ, Crofts TJ, Li AK. Endoscopic injection of adrenaline for actively bleeding ulcers: a randomised trial. Br Med J (Clin Res Ed) 1988 Jun 11;296(6637):1631–3. *[15] Sung JJ, Tsoi KK, Lai LH, Wu JC, Lau JY. Endoscopic clipping versus injection and thermo-coagulation in the treatment of non-variceal upper gastrointestinal bleeding: a meta-analysis. Gut 2007;56:1364–72. *[16] Marmo R, Rotondano G, Piscopo R, Bianco MA, D’Angella R, Cipoletta L. Dual therapy versus monotherapy in the endoscopic treatment of high risk bleeding ulcers: a meta-analysis of controlled trials. Am J Gastroenterol 2007; 102(2):279–89. [17] Saeed ZA. Second thoughts about second-look endoscopy for ulcer bleeding? Endoscopy 1998;30:650–2. [18] Villanueva C, Balanzo J, Torras X, et al. Value of second-look endoscopy after injection therapy for bleeding peptic ulcer: a prospective and randomized trial. Gastrointest Endosc 1994. [19] Messmann H, Schaller P, Andus T, et al. Effect of programmed endoscopic follow-up examinations on the rebleeding rate of gastric or duodenal ulcers treated by injection therapy: a prospective, randomized controlled trial. Endoscopy 1998;30: 583–9. 40:34–9. *[20] Saeed ZA, Cole RA, Ramirez FC, et al. Endoscopic retreatment after successful initial hemostasis prevents ulcer rebleeding: a prospective randomized trial. Endoscopy 1996;28:288–94. [21] Lin CK, Lai KH, Lo GH, et al. The value of second-look endoscopy after endoscopic injection therapy for bleeding peptic ulcer. Gastroenterology 1996;110:A177. *[22] Chiu PWY, Lam CYW, Lee SW, Kwong KH, Lam SH, Lee DTY, et al. Effect of scheduled second therapeutic endoscopy on peptic ulcer rebleeding: a prospective randomized trial. Gut 2003;52:1403–7. *[23] Marmo R, Rotondano G, Bianco MA, Pscopo R, Prisco A, Cipolletta L. Outcome of endoscopic treatment for peptic ulcer bleeding: Is a second look necessary? A meta-analysis. Gastrointest Endosc 2003;57:62–7. [24] Chiu PW, Lau TS, Kwong KH, Suen DT, Kwok SP. Impact of programmed second endoscopy with appropriate retreatment on peptic ulcer rebleeding: a systematic review. Ann Coll Surgeons Hong Kong 2003;7(4):106–15. [25] Tsoi KK, Chiu PW, Sung JJ. Endoscopy for upper gastrointestinal bleeding: is routine second-look necessary? Nat Rev Gastroenterol Hepatol 2009 Dec;6(12):717–22. [26] Chiu PW, Sung JJ. High risk ulcer bleeding: when is second-look endoscopy recommended? Clin Gastroenterol Hepatol 2010 Aug;8(8):651–4. [27] Chung SCS. Preventing ulcer rebleeding: the role of second-look endoscopy. Can J Gastroenterol 1999;13(5):409–11. [28] Tsoi KK, Chan HC, Chiu PW, Pau CY, Lau JY, Sung JJ. Second look endoscopi with thermal coagulation or injection for peptic ulcer bleeding: a meta-analysis. J Gastroenterol Hepatol 2010 Jan;25(1):8–13. *[29] Spiegel BM, Ofman JJ, Woods K, Vakil NB. Minimizing recurrent peptic ulcer hemorrhage after endoscopic hemostasis: the cost-effectiveness of competing strategies. Am J Gastroenterol 2003;98(1):86–97. *[30] Imperiale TF, Kong N. Second-look endoscopy for bleeding peptic ulcer disease: a decision-effectiveness and costeffectiveness analysis. J Clin Gastroenterol 2012;46(9):e71–5. [31] Barkun A, Bardou M, Kulpers E, Sng JJ, Hunt RH, Martel M, et al. International consensus recommendations on the management of patients with non-variceal upper gastrointestinal bleeding. Ann Intern Med 2010;152:101–13.
P.W.Y. Chiu / Best Practice & Research Clinical Gastroenterology 27 (2013) 905–911
911
[32] Sung JJ, Chan FK, Chen M, Ching JY, Ho KY, Kachinorn U, et al. Asia Pacific working group consensus on non-variceal upper gastrointestinal bleeding. Gut 2011;60:1170–7. [33] Lau JY, Sung JJ, Lee KK, Yung MY, Wong SK, Wu JC, et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers. N Engl J Med 2000 Aug 3;343(5):310–6.