An Audit of Argon Plasma Coagulation, Epinephrine Injection, and Proton-Pump Infusion Therapy in the Management of Bleeding Peptic Ulcer Disease

We read with interest the recent review by Rollhauser and Fleischer on nonvariceal upper gastrointestinal bleeding [1]. There is increasing evidence that dual therapy with injection and thermal treatment is becoming the gold standard for the treatment of high-risk bleeding peptic ulcers [1] [2]. Argon plasma coagulation may be an alternative thermal treatment, and indeed Cipolletta et al. [3] found that hemostasis was achieved significantly more quickly in actively bleeding ulcers when argon plasma coagulation was used in comparison with a heater probe, and that there was a trend toward fewer rebleeding episodes (15 % vs. 21 %). A recent study by Lau et al. demonstrated that a high-dose infusion of a proton-pump inhibitor (PPI) after endoscopic therapy reduces rebleeding rates [4].

At our institutions, we were interested in combining argon plasma coagulation with epinephrine injection and high-dose proton-pump infusion, as the paper by Cipolletta et al. does not include postprocedural PPI infusion. We present here an audit of our experience, comparing it with the previous form of management for this condition in our institutions, which was to use injected epinephrine with a standard-dose proton-pump inhibitor.

A retrospective cohort was identified via a computerized endoscopic database (EndoScribe, Health Communication Network, Adelaide, South Australia) between January 1999 and June 2001 (17 months), to provide a historical control group of 33 patients. The historical control group was enrolled using the same inclusion criteria as the prospective group. We then prospectively recruited 33 consecutive patients from June 2001 to September 2002 (total of 15 months). All patients presenting at Flinders Medical Center and Repatriation General Hospital with high-risk peptic ulcer disease were included. High-risk ulcers were defined as those with stigmata of a spurting vessel, ongoing oozing, or a nonbleeding visible vessel in the ulcer base. Endoscopy was carried out within 24 h of admission, after stabilization of the patient and admission, and subsequent endoscopic examinations were conducted in the intensive-care unit.

Rebleeding, mortality and surgery rates, transfusion requirements, and length of stay were compared with the 33 historical control patients who were treated prior to this trial with epinephrine injection alone followed by a standard-dose oral or intravenous proton-pump inhibitor, as decided by the treating gastroenterologist.

At endoscopy, epinephrine (diluted 1:10 000) was first injected in four quadrants in 0.5 - 1 ml aliquots to achieve hemostasis. If there was any clot overlying the ulcer, it was removed after the epinephrine injection. Argon plasma coagulation (APC) was then applied to the visible vessel. The power used for APC was operator-dependent, with a default current setting at 40 W, and the argon flow rate ranged from 0.4 to 2 l/min (with the latter being the default setting). In the event of rebleeding, a repeat endoscopic examination was performed with the aim of repeating endoscopic therapy if this was deemed appropriate. If endoscopic therapy was unsuccessful in achieving hemostasis or a second rebleed occurred, the patient was referred for surgery to be considered.

After successful endoscopic therapy, all patients were given 80 mg intravenous pantoprazole at once, followed by an intravenous pantoprazole infusion at 8 mg/h for 48 h. The intravenous infusion bags containing pantoprazole were changed every 3 h, in accordance with the manufacturer’s recommendations. All patients were then placed on 40 mg pantoprazole orally daily (unless Helicobacter eradication was first instituted) until a repeat endoscopic examination 6 - 8 weeks later.

There were 33 patients in each group. There were no significant statistical differences between them with regard to sex, age, and the size or location of the ulcers. With regard to ulcer stigmata, there tended to be more visible vessels and active bleeding in the prospective group than in the retrospective control group (P = 0.02). All of the patients completed the pantoprazole infusion treatment as per protocol.

The rebleeding rates were significantly lower in the combination therapy group (four of 33, 12.1 %, vs. nine of 33, 27.3 %; P = 0.025). The median transfusion requirement was not significantly reduced (two units, range 0 - 13, vs. three units, range 0 - 17; P = 0.244). The median length of the hospital stay was reduced (6 days, range 3 - 20 days, vs. 10 days, range 2 - 75; P = 0.023). The numbers were too small for meaningful comparison of the surgery and mortality rates (one vs. four undergoing surgery, one death in each group). In both groups, the deaths that occurred were unrelated to the endoscopy and medications used.

This small audit study demonstrates that combination therapy using epinephrine injection, argon plasma coagulation, and intravenous pantoprazole infusion is safe and effective. The rebleeding rate was reduced from 27 % to 12 %, with a significant decrease in the length of hospital stays. Intravenous pantoprazole was well tolerated via the infusion route, and initial minor thrombophlebitis did not recur once the pantoprazole was diluted. There were no complications of the argon plasma coagulation treatment.

Argon plasma coagulation has a number of theoretical advantages over heater-probe treatment. It is easier to reach difficult bleeding points, such as junction of the first and second parts of the duodenum, due to the arcing effect of the argon plasma. The ”no-touch” technique in APC, and the fact that the burn depth can be preset by the power setting chosen (e. g. a burn depth of 3 mm at the 40-W setting used in this study) should also reduce the risk of perforation in comparison with the heater probe. One theoretical disadvantage may be the lack of thermal coaptation of the kind that can be achieved using heater probes and bipolar diathermy techniques. However, the demonstration that APC appears to be effective indicates that thermal coaptation may not be as important as has previously been considered. These findings support the data published by Cipolletta et al. suggesting that argon plasma coagulation is a viable alternative to heater-probe treatment as a thermal modality, with a similar rebleeding rate. Whether this rebleeding rate would have been achieved without the intravenous pantoprazole infusion remains unclear.

Argon plasma coagulation and intravenous pantoprazole administration as part of the combined therapy described here was thus effective and warrants further evaluation as part of a larger randomized and controlled trial.

References

• 1 Rollhauser C, Fleischer P E. Nonvariceal upper gastrointestinal bleeding.  Endoscopy. 2002;  34 111-118
• 2 Elta G H. Acute Nonvariceal upper gastrointestinal hemorrhage.  Curr Treat Options Gastroenterol. 2002;  5 147-152
• 3 Cipolletta L, Bianco M A, Rotondano G. et al . Prospective comparison of argon plasma coagulator and heater probe in the endoscopic treatment of major peptic ulcer bleeding.  Gastrointest Endosc. 1998;  48 191-195
• 4 Lau J Y, Sung J J, Lee K K. et al . Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers.  N Engl J Med. 2000;  343 310-316