Upper GI bleed: balloon tamponade controlled bleeding faster, but caused more complications

Clinical bottom line (level 1b)

  1. Patients with bleeding oesophageal varices who had a Linton-Michel haemostatic balloon inserted were more likely have bleeding controlled within 24 hours (NNT = 4 at 24 hours) , but not at one week (NNH = 4 at 7 days) than terlipressin
  2. There was no clear effect on mortality.
  3. Patients who had a Linton-Michel balloon inserted rather than i/v terlipressin were at increased risk of complications (NNH = 2 at 7 days)
Garcia-Compean et al: Archives of Medical Research 1997; 28: 241-245
Expires October 2002

The study

Unblinded ?concealed randomised trial without intention-to-treat
Setting: acute hospital, France

40 patients (aged mean 53, 65% male) with cirrhosis and bleeding oesophageal varices

Excluded if
  • aged < 17 or >75
  • coronary insufficiency, myocardial infarction, uncontrolled arterial hypertension, cardiac arrhythmia
  • peripheral vascular insufficiency
  • severe chronic renal insufficiency

Control Group: (n = 20, 20 analysed): terlipressin iv 1-2 mg depending on body weight every 4 hours
Experimental Group: (n = 20, 20 analysed): Linton-Michel tube for 24 hours

100% followed for 7 days
Outcome notes:
  • failure to control bleeding control : transfusion of 4 units of blood over 24 hours required
  • complications : gastroesophageal gastrooesophageal ulcers, arterial hypertension, thoracic pain

The evidence

Outcome Time to outcome CEREERRRR
(95% CI)		ARR
(95% CI)		NNT
(95% CI)
failure to control bleeding control 24 hours 6
(30.0%)		 1
(5.00%)		 83%
(-26% to 98%)		 25%
(2.76% to 47.2%)		 4
(2 to 36)
rebleeding 7 days 2
(10.0%)		 7
(35.0%)		 -250%
(-1383% to 17%)		 -25.0%
(-49.7% to -0.30%)		 -4
(-328 to -2)
death 7 days 4
(20.0%)		 6
(30.0%)		 -50%
(-350% to 50%)		 -10.0%
(-36.7% to 16.7%)		 -10
(NNT = 3 to infinity;
NNH = 6 to infinity)
complications weeks 3
(15.0%)		 13
(65.0%)		 -333%
(-1191% to -45%)		 -50.0%
(-76.1% to -23.9%)		 -2
(-4 to -1)

Comments

  1. Treatment with terlipressin is easier but less effective than balloon haemostasis - skill of operator should be taken into account.

Citation

  1. Garcia-Compean D, Blanc P, Bories J-M, et al: Treatment of active gastroesophageal variceal bleeding with terlipressin or hemostatic balloon in patients with cirrhosis. A randomized controlled trial. Archives of Medical Research 1997; 28: 241-245
Contributor: Chris Ball and Musab Hayatli, October 1999
Reviewer: Ali Taha

Clinical Question.
Patient bleeding esophageal varices, oesophageal varices
Intervention or Exposure terlepresin, Linton-Michel haemostatic balloon
Outcome bleeding control, death suppression