This is Journal Watch, a twice-weekly survey
of new medical research, produced by the
Massachusetts Medical Society, publishers
of The New England Journal of Medicine,
Health News, AIDS Clinical Care and Clinical
Care for Prostatic Diseases.
  Twice a week, our physician-editors summarize
clinical research from a group of about 25
journals. We post the top two summaries,
along with the complete table of contents,
to selected news groups in sci.med.
  To receive the complete complement of stories
*via email* please send your snail-mail address

mailing list immediately and mail you an invoice
(the cost is $US50 a year).
                    *********
  Contents copyright 1995, Mass. Medical Society.

Journal Watch Summaries for Friday, April 28, 1995

INCREASED SURGICAL BLEEDING WITH A CALCIUM BLOCKER.
  BMJ 1995 Mar 25; 310:776-7.

CAN PROPHYLACTIC ASPIRIN CAUSE BLEEDING ULCER?.
  BMJ 1995 Apr 1; 310:827-30.

ERYTHROPOIETIN FOR PREDIALYSIS RENAL PATIENTS.
  Am J Kidney Dis 1995 Apr; 25:548-54.

PHYSICAL FITNESS IMPROVES LONGEVITY.
  JAMA 1995 Apr 12; 273:1093-8.

FRUITS AND VEGETABLES LINKED TO LOWER STROKE RISK.
  JAMA 1995 Apr 12; 273:1113-7.

DOES FISH INTAKE REDUCE CORONARY RISK?.
  N Engl J Med 1995 Apr 13; 332:977-82.

GREEN TEA: ANOTHER CARDIOPROTECTIVE AGENT?.
  BMJ 1995 Mar 18; 310:693-6.

DT.-950428.

AN.-1.
HL.-INCREASED SURGICAL BLEEDING WITH A CALCIUM BLOCKER.
NS.-Although calcium antagonists have not been linked with
an increased risk for post-surgical bleeding, there are
theoretical reasons to expect an association: these {*filter*}
are vasodilators and have antiplatelet action. Now, an
unexpected finding -- from a trial designed with another
purpose -- raises some concern.
   Believing that a calcium blocker might prevent neurologic
complications of valve-replacement surgery, investigators
randomized patients to nimodipine or placebo, administered
12 hours before cardiac valve replacement and four times
daily for five days thereafter. They planned to study 400
patients, but stopped at 149 because of significantly higher
mortality in the nimodipine group (11 vs. 1 percent).
   The two groups were similar when assessed for 50
preoperative variables. The major factor associated with the
excess mortality was postoperative bleeding: 13 percent of
the nimodipine group, versus 3 percent of controls, required
more than 10 units of {*filter*} or had early postoperative
chest-tube drainage exceeding 2400 ml. Five of the 8 deaths
in the nimodipine group involved major bleeds: 3 resulted
directly from {*filter*} loss, and 2 from major organ failure
caused by {*filter*} loss. Bleeding was a possible contributor in
2 of the 3 remaining deaths.
   Comment: This report represents a post-hoc analysis done
because of a surprise finding, and subsequent studies may
not bear it out. But in the meantime, clinicians should take
care in using calcium blockers in patients undergoing valve
replacement surgery. --KI Marton.
AU.-Wagenknecht LE; et al.
TI.-Surgical bleeding: unexpected effect of a calcium
antagonist.
SO.-BMJ 1995 Mar 25; 310:776-7.

AN.-2.
HL.-CAN PROPHYLACTIC ASPIRIN CAUSE BLEEDING ULCER?.
NS.-Most studies that have linked aspirin use with
bleeding from peptic ulcer have involved doses above 300
mg/day. Many people take lower doses for cardiovascular
prophylaxis, but prospective data on the associated bleeding
risk are inconclusive.
   A British case-control study recruited 1121 patients with
gastric or duodenal ulcer bleeding and two groups of age-
and sex-matched controls: 1126 admitted to the same
hospitals without bleeding, and 989 from the community.
Regular aspirin use (at least five times weekly for at least
the past month) was reported by 12.8 percent of the cases, 9
percent of hospital controls, and 7.8 percent of community
controls. After correcting for prior ulcer or dyspepsia,
concurrent NSAID use, smoking, and {*filter*} intake, odds
ratios for bleeding were increased at all daily aspirin
doses: 2.3 for 75 mg, 3.2 for 150 mg, and 3.9 for 300 mg.
The risk was highest for Alka-Seltzer (odds ratio, 7.7) and
lowest for enteric coated aspirin (1.1).
   A randomized, crossover study indirectly supports these
findings. Researchers in Philadelphia administered 325
mg/day of plain and enteric coated aspirin, in random order,
to 17 healthy young {*filter*}s. Mean daily fecal {*filter*} loss
(measured by chromium-51 red-cell labeling) was 0.47 ml at
baseline, 0.96 ml after a week of enteric coated aspirin,
and 1.82 ml after a week of plain aspirin. All comparisons
were significant. A two-month study with another 30 subjects
yielded similar findings.
    Comment: These data support the contention that even low
aspirin doses can increase the risk for GI bleeding. The
authors of the British study appropriately note, however,
that the benefits of low-dose aspirin probably still
outweigh the detriments. Both studies support the use of
enteric coated aspirin for cardiovascular prophylaxis.
--KI Marton and AS Brett.
AU.-Weil J; et al.
TI.-Prophylactic aspirin and risk of peptic ulcer
bleeding.
SO.-BMJ 1995 Apr 1; 310:827-30.
AU.-Savon JJ; et al.
TI.-Gastrointestinal {*filter*} loss with low dose (325 mg)
plain and enteric coated aspirin administration.
SO.-Am J Gastroenterol 1995 Apr; 90:581-5.