Pahor M, Guralnik JM, Furberg CD, Carbonin P, Havlik RJ
Risk of gastrointestinal hemorrhage with calcium
antagonists in hypertensive persons over 67 years old
Lancet
(Apr) 347:1061-1065 1996

It is known that some calcium channel blockers may increase the
bleeding time via an effect on platelets. Pahor et al used data from
a prospective cohort study of elderly patients (greater than 65 years
of age) living in the Boston area and Iowa, to assess the risks of
gastrointestinal (GI) hemorrhage as factored by type of anti-
hypertensive medication taken.
At baseline, of 1636 hypertensive patients, 851 were taking beta-
blockers alone, 21 were taking ACE-inhibitors, and 137 Ca-channel
blockers. Patients taking combinations of these drugs were excluded.
After 3 years of follow-up, 557 were on beta blockers, 204 on ACE
inhibitors, and 334 on calcium antagonists. The risk of severe GI
bleeding was higher in patients taking calcium channel blockers (RR
1.68, confidence interval 1.03-2.74) than in those taking beta
blockers. Interestingly, in this population, the relative risk of
gastroduodenal ulcer and gastritis was higher in the calcium channel
blocker group as well, a possible confounder. Verapamil and diltiazem
were associated with the higher GI bleeding risk, whereas nifedipine
was not.
Comment: The study is interesting and plausible, although a
number of potential confounding factors may have been present. The
higher incidence of ulcer and gastritis in the patients treated with
Ca-channel blockers points to an alternative explanation for the
results. Interestingly, the relative risk for cancer of any kind was
2.98 with calcium channel blockers, with a 95% confidence interval of
1.2-7.6, although nothing was made of this. Perhaps this will be the
next paper by these authors. Clearly it seems to be open season on
calcium channel blockers. This study clearly needs to be confirmed
before it seriously is used to alter clinical practice. I believe
that the recommendation to use calcium channel blockers in elderly
patients at risk for GI
bleeding is premature based on the
data presented.
(John T. Daugirdas, M.D., University of Illinois at Chicago)