Prevention of First Bleeding in Cirrhosis
Clinical Question
Will primary prophylaxis with oral B-blockers reduce the risk of initial
bleeding from esophageal varices in patients with cirrhosis?
Clinical Bottom Lines
- ß-blockers are indicated for the prevention of first bleeding from
esophageal varices in patients with cirrhosis.
- The effectiveness of sclerotherapy for prevention of first bleeding from
esophageal varices remains to be determined.
The Evidence
- Pooled Odds Ratio (POR) for bleeding among patients treated with ß-blockers
vs no treatment was 0.54 (95%CI 0.39 - 0.74)*.
- The number needed to treat (NNT) to prevent one bleeding episode is 11
(95%CI 8 - 18)**.
- POR for prevention of death was 0.75 (95%CI 0.57 - 1.06).
- POR for bleeding in patients treated with sclerotherapy vs no treatment was0.60
(95%CI 0.49 - 0.74)***.
Comments
- This was a meta-analysis of 9 to 19 randomized, double-blinded, controlled
trials addressing prevention of first episodes of bleeding in patients with
esophageal variceal bleeding secondary to cirrhosis.
- The statistical analyses were extremely rigorous.
- *In this group, one outlier trial could be removed from the overall
analysis secondary to statistically significant heterogeneity, resulting in a
POR of 0.48 (95%CI 0.35 - 0.66). As can be seen, adding in this trial did not
change the results significantly.
- **The NNT was likely changed, while removing the outlier, the NNT is only
9 (CI 5 - 12).
- ***Although they reported this for completeness, they demonstrated a
significant heterogeneity among all the trials for this outcome, thus
making the statistically significant POR meaningless.
- The ability of a meta-analysis allowed the results of all those trials to
be combined, resulting in 996 patients. This study is valid by all JAMA
criteria, and the results can be applied to your patients.
- In addition, the results of this study cannot and must not
be extrapolated to those patients with recurrent variceal bleeding.
APPRAISED BY: Patrick Sousa, MD
DATE: November 15, 1995
Prevention of First Bleeding in Cirrhosis; A Meta-Analysis of
Randomized Trials of Nonsurgical Treatment. Pagliaro, et al. Ann Int Med.
July 1992;117:59-70.

