Send letter to journal:
Re: Cost-effectiveness of adding PPI to scheduled second endoscopy

Dear Editor

We appreciated the important issue raised by Kayhan and colleagues concerning adjunct PPI infusion.[1] Lau et al published a landmark prospective randomized trial comparing adjunct PPI infusion against placebo after therapeutic endoscopy for bleeding peptic ulcers.[2] They showed that with the use of adjunct PPI infusion, the rate of rebleeding is reduced to 6.7% compared to 22.5% in those treated with placebo alone. The cost-effectiveness of this strategy, however, is of a major concern if we added PPI infusion together with a scheduled second endoscopy. With the performance of scheduled second endoscopy, we can achieve a 5% rebleeding which is comparable to that achieved by adjunct PPI infusion.[3] It would be very difficult to attain a big enough sample size to conduct a trial comparing addition of adjunct PPI infusion to scheduled second endoscopy versus PPI infusion alone as suggested. We will need a sample size of more than 1000 patients in each arm in order to show a reduction in rebleeding rate from 6.7% to 1%.

Perhaps a study comparing the cost-effectiveness of the two major strategies of prevention of rebleeding is more appropriate, ie the use of adjunct PPI infusion or scheduled second endoscopy. Indirect evidence from Spiegel et al suggested that adjunct PPI infusion could be the most cost- effective strategy in preventing rebleeding.[4] However the analysis was performed before the publication of our prospective randomized trial. We believe that the issue of cost-effectiveness is still unresolved.

References

1. Kayhan B, Akbulut S, Özaslan E. Not enough PPI infusion [electronic response to Chiu et al. Effect of scheduled second therapeutic endoscopy on peptic ulcer rebleeding: a prospective randomised trial] gutjnl.com 2003 URL direct link to eLetter

2. Lau JYW, Sung JJY, Lee KKC et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers. New England Journal of Medicine 2000; 343(5): 310-316

3. Chiu PWY, Lam CYW, Lee SW et al. Effect of scheduled second endoscopy on peptic ulcer rebleeding: a prospective randomized trial. Gut 2003; 52(10): 1403-1407.

Send letter to journal:
Re: Not enough PPI infusion

Dear Editor

I read the recently published article by Chiu and associates,[1] with interest. Although the aim of this study is very interesting, we think not all patients received enough PPI infusion after the first endoscopic treatment. If all the patients (both groups; control and study groups) were given an infusion of 8 mg/hr for 72 hours,[2] the results may be different.

References

1. Chiu PWY, Lam CYW, Lee SW, Kwong KH, Lam SH, LeeDTY, Kwok SPY. Effect of scheduled second therapeutic endoscopy on peptic ulcer rebleeding: a prospective randomised trial. Gut 2003;52:1403-07.

Send letter to journal:
Re: Author's reply

Dear Editor,

I appreciated the letters from Professor Richard G Fiddian-Green. There are numerous meaningful thoughts that worth further discussion.

Firstly, as Professor Fiddian-Green rightly pointed out, randomized trials performed in the 1980s confirmed that simple diagnostic endoscopy did not improve outcomes for acute non-variceal upper gastrointestinal bleeding (1,2). Within the last 15 years, however, more than 30 randomized trials and two meta-analysis did show that by performing a primary therapeutic endoscopy, we could achieve a significant reduction in the rates of mortality, recurrent bleeding and surgery (3,4). These randomized trials compared a group of patients that underwent primary endoscopic therapy against a control, which was treated by salvage surgery once re- bleeding occurred. In this perspective, therapeutic endoscopy did showed an improvement in the outcomes when compared with salvage surgery.

We also noticed that the rate of operative intervention decreased from 25% before 1980s to 12% recently (6,7). With scheduled second endoscopy, we could further reduce the rate of surgical intervention (8). On the contrary, those that required surgical intervention were older, having more co-morbidities and higher postoperative mortality. The reported operative mortality was up to 30%. Although therapeutic endoscopy successfully reduced the number of patients that required surgical intervention, relying too much on endoscopic therapy might jeopardize a small but important amount of patients that eventually needed surgical hemostasis.

Before the era of endoscopy therapy, the only method of hemostasis and prevention of rebleeding was surgery. There was a huge debate as to whether an aggressive surgical approach should be adopted. Morris et al compared early against delayed surgery for treating bleeding peptic ulcers. They found that patients older than 60 years had a significantly higher mortality in the delayed group as compared to those in the early surgery group (9). This indirectly suggested that early surgical intervention might improve clinical outcomes. Other authors, however, disagreed by pointing out that a lower rate of surgical intervention correlated with a lower mortality (10,11). Dronfield et al compared different operative policies between two hospitals in the same region. A significantly higher operative rate was noticed in one hospital (32% vs 46%, p = 0.03). However, the hospital with a lower operative rate had a lower overall mortality (17.7% vs 12.7%). Thus whether an aggressive surgical approach might lead to a better outcome is yet to be resolved.

Currently, endoscopic therapy remained the most effective method of primary hemostasis for bleeding peptic ulcers, while performing a scheduled second endoscopy identify and treat those high risk ulcers before they re-bleed. With these advances in endoscopy, the remaining ulcers that re-bleed perhaps represented a selected group that is difficult and nasty. The next step forward is to identify those that have a high risk of recurrent bleeding not amendable by endoscopic therapy and intervene early with surgery. The common reported predictive factors for recurrent bleeding included age, ulcer size, shock on admission and anemia (12). Professor Fiddian-Green suggested an innovative idea of measuring gastric intramucosal acidosis as an indicator for visceral hypoperfusion and assisted in early detection of those that required surgery. Whether gastric intramucosal acidosis can predict the need of surgical intervention is an important clinical question and warrants further investigation.

Endoscopic therapy has its own limitation. Johnston et al studied the effect of hemostasis using BICAP and heat probe on various diameter of artery in a canine model (13). They found that the percentage of coagulation dropped significantly when the size of the artery exceeded 2 mm. Hence for those with ulcers that bled from a vessel with a diameter greater than 2 mm, endoscopic therapy is unlikely to work. The only practical issue is how to identify those patients that have a bleeding vessel of 2 mm or above, and treat them with surgery to prevent a catastrophic re-bleeding. The size of the bleeding vessel served as another important indicator for surgical intervention, and this warrant further investigation on how to identify those bleeding vessels greater than 2 mm. We believe the current development of endoscopic treatment for bleeding peptic ulcers is a big step forward. The role of early surgical intervention in those patients with high risk of re-bleeding is uncertain. Future advance in the treatment of bleeding peptic ulcers is not only about how to do well, but towards the path of doing the right thing at the right time.

References

1.Petersen WL, Barnett CC, Smith HJ et al. Routine early endoscopy in upper gastrointestinal tract bleeding. A randomized controlled trial. New England Journal of Medicine 1981;304:925-929.

2.Graham DY. Limited value of early endoscopy in the management of acute upper gastrointestinal bleeding. American Journal of Surgery 1980;140:284-290.

3.Sacks SH, Chalmers TC, Blum AL, et al. Endoscopic hemostasis ¡V an effective therapy for bleeding peptic ulcers. JAMA 1990; 264(4):494-499.

4.Cook DJ, Guyatt GH, Salena BJ, et al. Endoscopic therapy for acute non-variceal upper gastrointestinal hemorrhage: A Meta-analysis. Gastroenterology 1992; 102:139-148.

5.Rockall TA, Logan RFA, Devlin HB, et al. Incidence of and mortality from acute upper gastrointestinal haemorrhage in the United Kingdom. British Medical Journal 1995; 311:222-226

6.Vellacott KD, Dronfield MW, Atkinson M, et al. Comparison of surgical and medical management of bleeding peptic ulcers. British Medical Journal 1984; 288:1277-1280.

7.Rockall TA. Management and outcome of patients undergoing surgery after acute upper gastrointestinal haemorrhage. Journal of the Royal Society of Medicine 1998;91(10): 518-523.

8.Chiu PWY, Lam CYW, Lee SW, et al. Effect of scheduled second therapeutic endoscopy on peptic ulcer rebleeding: a prospective randomized trial. GUT 2003;52: 1403-1407.

9.Morris DL, Hawker PC, Brearley S, et al. Optimal timing of operation for bleeding peptic ulcer: prospective randomized trial. British Medical Journal 1984; 288: 1277-1280.

10.Dronfield MW, Atkinson M, Langman MJS. Effect of different operation policies on mortality from bleeding peptic ulcer. Lancet 1979; 1:1126-1128.

11.Rofe SB, Duggan JM, Smith ER, et al. Conservative treatment of gastrointestinal haemorrhage. GUT 1985; 26:481-484.

12.Wong SKH, Yu LM, Lau JYW, et al Prediction of therapeutic failure after adrenaline injection plus heater probe treatment in patients with bleeding peptic ulcer. GUT 2002; 50:322-325.

Send letter to journal:
Re: Preventing rebleeding from PUD

Dear Editor

It is virtually impossible to stop permanently massive bleeding from a large artery, such as a gastroduodenal of left gastric, with injections of epinephrine, clips or sutures if they could be placed endoscopically or even laparoscopically. If therapeutic endoscopy is effective, as many claim, then it is most likely to be effective for bleeding from small arteries which tend to stop spontaneously unless a patient is rendered hypertensive or even normotensive by resuscitation with intravenous fluids and/or blood products. [For this reason many surgeons will not bother to stop subcutaneous bleeding after making their incision and simply have their assistants place a swab or pack on the bleeding sites . Some will even leave them there to protect the skin during retraction until the end of the operation when the bleeding will have stopped].

The rebleeding rate after definitive surgery to control bleeding from peptic ulcers is high and the accompanying risk of dying is greatly increased. It is particularly high in the hands of the novice gastric surgeon. There are several causes for the rebleeding. First is the failure to obtain haemostasis in the relevant pathology in patients with more than one potential cause [peptic ulceration coexiting with varices being a notable example]. Second is the failure to deal with the bleeding site in an appropriate manner. The third is the appearance of new bleeding sites, most commonly a reactionary haemorrhage from a suture line or stress ulceration. Fourth is the development of a coagulopathy.

The failure of therapeutic endoscopy may leave one or more "suction" injuries which may or may not bleed. The more aggressive the attempts to stop the bleeding the more likely the creation of these red herrings. These may also be caused by Ewald tubes and in ICU patients with fragile mucosa by nasogastric tubes. In any event the value of compulsive irrigation and suctioning of the stomach is highly questionable for in those who require surgery the stomach is often filled with large clots that cannot be sucked out even with the largest of Ewald tubes and may have to be scooped manually to empty the stomach

Arterial bleeders will often have stopped by the time the ulcer is exposed at surgery because the artery has gone into spasm and may even have been temporarily occluded by clot. This is a particularly common cause of failure to obtain proper haemostasis from a large bleeding artery. Hypotension at the time is no doubt a contributing factor. It is, therefore, advisable to "kick" the ulcer after under-running the bleeder to be absolutely sure that the bleeding site has been permananently secured.

It is very easy for the novice surgeon to oversew the ulcer without securing the bleeder at all rather than under-running the bleeding vessel. If this is done even "kicking" it might not reveal the technical failure for it is hidden under apposed mucosa. To under-run the gastroduodenal artery it is vital that the ulcer be opened to expose the vessel in its base by Kocherising the duodenum and lifting the ulcer with a finger tip applied to the back of the ulcer from dorsum of the pancreas. If this is not done the ulcer folds in on itself making under-running impossible and oversewing the only physical option. The best way to secure bleeding from a chronic gastric ulcer is to perform a gastrectomy. Lesser procedures are awfully messy and difficult to accomplish reliably especially for the novice.

Staplers are not haemostatic and their anastomostic lines need to be oversewn and invaginated as in the construction of a hand-sewn anastomoses if reactionary haemorrhage from the anastomoses is to be reliably avoided. In those anastomoses constructed by hand-sewing alone having an assistant who knows just how much tension to apply in "following" is critical in avoiding reactionary haemorrhage. As my mentor Professor Jannie Louw at Groote Schuur always said when I assisted him, "if the patient bleeds after surgery its your fault". [He never used staplers].

The best way of avoiding bleeding from a coagulopathy, injury to fragile mucosa and rebleeding from stress ulceration is to send the patient to surgery very early in the course of the disease so that the duration of gastric mucosal ischaemia will have be limited. A fall in gastric intramucosal pH is the best predictor of bleeding from stress ulcers in a patient with a given number of "risk factors".[1] [Most of these risk factors are organ dysfunctions and failures which have developed over a week or two. Bleeding from stress ulceration tends to occur in the second week after the onset of an acute illness rather than earlier in the course of the disease even though mucosal erosions might have been visualised endoscopically at that time].

If the need for surgery can be determined by the rate of bleeding derived from repeated or conceivably continuous measurements of blood volume, as proposed in my earlier electronic communication, then an accurate measure of the rate of bleeding could be obtained in as little as ten to fifteen minutes after introducing the labelled erythrocytes.[2] If the pattern of bleeding is a disciminating factor [massive bleeding from an eroded artery being characteristically intermittent] then identification of this pattern should be possible within an hour or two of admission.

The keys to achieving a lower mortality in patients with gastrointestinal haemorrhage would seem to be to limit the duration of an intramucosal acidosis to less than an hour if at all , to decide within that hour whether the patient requires surgery to stop the bleeding, and to take the patient to surgery immediately without administering wither fluids or blood products as soon as possible.[3] Repletion of the blood volume might be best deferred until after the bleeding vessel has been permanently secured and even left to physiolgical compensatory mechanisms as the late Marvin Pollard advocated to his bewildered residents.

References

(1) Fiddian-Green RG, McGough E, Pittenger G, Rothman E. Predictive value of intramural pH and other risk factors for massive bleeding from stress ulceration. Gastroenterology. 1983 Sep;85(3):613-20.

(2) Beckman RL, Pittenger GL, Swanson DP, Thrall JH, Fiddian-Green RG. Blood loss measured with indium-111-labeled red blood cells in dogs. Radiology. 1983 Jul;148(1):243-5.

(3) Bickell WH, Wall MJ Jr, Pepe PE, Martin RR, Ginger VF, Allen MK, Mattox KL. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med. 1994 Oct 27;331(17):1105-9.

Send letter to journal:
Re: Moving forwards not sideways

Dear Editor

I am not aware of any data that have shown beyond doubt that therapeutic endoscopy improves outcome. This study showed that second therapeutic endoscopy failed to improve outcome but reduced the number of cases "requiring" surgery. This raises serious questions about the validity of current indications for surgery for operative mortality is high in these circumstances and has not improved in almost five decades. The mortality m,ight even have increased because surgeons are less experienced with ulcer surgery, those dealing with bleeding ulcers being particularly challenging.

It is now known that the development of shock, defined as a gastric intramucosal acidosis, is the best predictor of outcome in the critically ill the likelihood of developing organs dysfunctions and dying increasing as the degree and duration increases.[1] More importantly in this context is that the presence of haemodynamically compensated shock before surgery has an adverse effect upon outcome.[2] The goal should, therefore, be to identify operative candidates within one hour of their development of a gastric intramucosal acidosis and operate immediately on those who require surgery. The question is how to do this without subjecting patients to unnecessary operations.

One way of doing this might be to use sequential measurements of the magnitude of blood loss on the assumption that those most likely to benefit from surgery will be those se rate of blood loss is greatest. One way of obtaining sequential measurements of the rate of blood loss would be to use indium-111-labeled red blood cells or surrogate.[3]

The first step should be to perform a prospective observational study with organ dysfunctions, thirty-day and possibly six-month mortalities as end-points. Better citeria for selecting patients who require surgery might be derived from the analysis of these data and tested in a complementary rospective interventional study.

References

(1) Fiddian-Green RG, Haglund U, Gutierrez G, Shoemaker WC. Goals for the resuscitation of shock. Crit Care Med. 1993 Feb;21(2 Suppl):S25-31.

(2) Poeze M, Takala J, Greve JW, Ramsay G Pre-operative tonometry is predictive for mortality and morbidity in high-risk surgical patients. Intensive Care Med. 2000 Sep;26(9):1272-81.

(3) Beckman RL, Pittenger GL, Swanson DP, Thrall JH, Fiddian-Green RG. Blood loss measured with indium-111-labeled red blood cells in dogs. Radiology. 1983 Jul;148(1):243-5.

Send letter to journal:
Re: Diagnotic endoscopy: does it help?

Dear Editor

Having questioned the value of therapeutic endoscopy let me question the value of diagnostic endoscopy having lived through the days from when it was not available to its current status. Having given the matter considerable thought[1] I seriously doubt that it has been of any value. It might even have had an adverse effect on outcome but delaying operative intervention in those who need it and amongst whom most deaths occur.

If done emergently it is difficult to visualise the bleeding site especially in those who require surgery and whose rate of blood loss is greatest. Its greatest value may be in excluding the presence of oesophaeal bleeding from varicies but this is arguably best done on the operating table by the surgeon. In the UK many centers no longer perform endoscopy emergently preferring to wait until the bleeding has stopped and a better diagnostic evaluation can be made on the daily endoscopy list. This may be good for documentation but is of doubtful benefit in managing patients because of the inevitable delay in surgical intervention in those that need it.

Operative mortality for bleeding ulcers used to be in the order of 15%. I suspect it is very much higher today because surgeons do far less ulcer surgery. This is likely to be especially true if the operations are delegated to registrars in the middle of the night. It might be wise to take a fresh look at the problem.