Intended for healthcare professionals

Letters

Adding heat probe treatment to adrenaline injection for spurting haemorrhage of peptic ulcers

BMJ 1997; 315 doi: https://doi.org/10.1136/bmj.315.7114.1016 (Published 18 October 1997) Cite this as: BMJ 1997;315:1016

Injection of adrenaline and human thrombin is best option

  1. S H Hussaini, Lecturer in medicinea,
  2. MA Hull, Lecturer in medicinea
  1. a Division of Medicine, St James's University Hospital, Leeds LS9 7TF
  2. b Chinese University of Hong Kong, Shatin, New Territories, Hong Kong

    Editor—Chung et al reported a randomised study examining the efficacy of giving heat probe treatment after endoscopic adrenaline injection for actively bleeding peptic ulcers.1 No benefit over adrenaline injection alone was detected. However, a subgroup analysis of patients with spurting haemorrhage showed benefit (reduced surgical rates) in the group given dual treatment. We are concerned about the stratification of patients in this subgroup analysis.

    Endoscopic treatment is more likely to fail in patients with posterior duodenal ulcers than in patients with anterior ulcers.2 3 Although the distribution of ulcer sites did not differ overall between the two treatment groups, it would be relevant to know the distribution of ulcers in the subgroup with spurting haemorrhage. For example, a greater number of posterior ulcers in the group given adrenaline injection alone would favour an apparent beneficial effect of dual treatment. Furthermore, comorbidity influences rebleeding rates.4 Thus more data on comorbidity should have been provided, with stratification in the subgroup analysis for this variable.

    Only two of the four end points used in the study were significant, some of the ulcers in the patients given dual treatment perforated, and the patients were incompletely stratified. These factors cast some doubt on the “advantageous” role of combined adrenaline injection and heat probe treatment in spurting haemorrhage. For the time being, endoscopic injection of adrenaline and human thrombin—an alternative therapeutic option not discussed by Chung et al—may be a more sensible option, since this has been shown to reduce ulcer rebleeding and mortality.5

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.

    Authors' reply

    1. S C Sydney Chung, Professor, department of surgeryb,
    2. James Y W Lau, Senior medical officer, department of surgeryb,
    3. Arthur K C Li, Vice chancellorb
    1. a Division of Medicine, St James's University Hospital, Leeds LS9 7TF
    2. b Chinese University of Hong Kong, Shatin, New Territories, Hong Kong

      Editor—Posterior duodenal ulcers are closer to large vessels than anterior duodenal ulcers are and are more likely to rebleed. Whether one can accurately identify such ulcers at endoscopy, however, is doubtful. In a study by Straker et al, true posterior location was identified in only 30% of cases by an experienced endoscopist.1 In our experience, there is often a considerable discrepancy between the location of duodenal ulcers reported at endoscopy and their true location at laparotomy. We therefore chose not to stratify our patients on the basis of the apparent location of the ulcers.

      Several studies have indicated that dual treatment is superior to injection of adrenaline alone. Kubba et al obtained excellent results by combining human thrombin and adrenaline.2 In their study a single operator did all the endoscopic treatment. Excellent results obtained by enthusiasts need to be validated by other centres before the technique is generally adopted. The possibility that human blood products might transmit disease, however remote, must also be taken into consideration.

      References

      1. 1.
      2. 2.
      View Abstract