Secondary Prophylaxis of Esophageal Variceal Bleeding by Nadolol Plus Isosorbide Mononitrate

Clinical Question

Is treatment with combined B-blocker and nitrate of patients with bleeding esophageal varices more effective than sclerotherapy for secondary prophylaxis of rebleeding?

Clinical Bottom Line

  1. Therapy with Nadolol (a ß-blocker) and ISMN reduces rebleeding episodes in patients with esophageal varices compared to sclerotherapy, but there is significant overlap of the values between groups.
  2. This combination probably does not afford any increase in efficacy compared with B-blockade alone.

The Evidence

Nadolol + ISMN (± SD) Sclerotherapy (± SD) p-value
Rebleeding episodes/patient 0.3 ± 0.6 1.1 ± 1.2 < .001
Rebleeding index* 15.5 ± 7.6 12.1 ± 9.7 = .025
Transfusion requirement (units) 2.6 (range 0 - 9) 4.9 (range 0 - 12) < .001

*Calculated as the number of months of followup divided by the number of rebleeding episodes per patient.


Actuarial Probability curves of remaining free of bleeding at 24 months showed a 20% difference between groups, with the sclerotherapy group having more rebleeding than the drug group. No confidence intervals are provided, but the p-value was 0.002

Comments

  1. This was a prospective, randomized study, but several flaws warrant discussion.
  2. The study was not blinded.
  3. There was no "true" control group (a group getting a "placebo" therapy.) However, the therapy under investigation was being compared to sclerotherapy which is serving as the control group. There have been multiple studies and meta-analyses assessing the efficacy of B-blockers in general in secondary prophylaxis of esophageal variceal bleeding. This study compares combination therapy to sclerotherapy, however.
  4. It would have been more useful to compare Nadolol + ISMN to Nadolol or ISMN alone.
  5. They employed an intention-to-treat analysis.
  6. No confidence intervals were calculated, but based simply on SD's, there is significant overlap between groups with respect to rebleeding, despite there being significant p-values.
  7. Based on the rebleeding index, the ARR range would be 0.055 to 0.013, giving a range of NNT (number needed to treat) of 20 - 100. Again, this is similar to previously published studies. There does not appear to be a significant benefit to adding ISMN to B-blockade alone, if you compare these numbers to previous studies.
  8. Based on actuarial curves, there is no difference in the death rate between groups.

APPRAISED BY: Katy Wong, MD

DATE: September 19,1996

Villanueva C, et al. Nadolol Plus Isosorbide Mononitrate Compared With Sclerotherapy for the Prevention of Variceal Rebleeding. NEJM 1996;334:1624-1629.