| |
Combination
therapy best
for high-risk ulcer bleeds
By
Diana Swift
NEW
ORLEANS, LA--There
is little evidence to suggest that emergency or early endoscopy
actually saves lives in upper GI bleeding, according to findings
from the Cleveland Clinic research group.
The group looked at 909 consecutive upper GI bleed patients from
13 centres, 64 per cent of whom had early (but not emergency) endoscopy
versus standard endoscopy. "In patients in the high-risk group-defined
as having actively bleeding spurters or ulcers-endoscopy within
the first 24 hours reduced rebleed rates and surgery requirements
and substantially shortened length of stay," Joseph Sung, MD,
said. The benefits were not apparent, however, in those having moderate
risk of ulcer rebleeding or clean-based ulcers. "This study
did not report on mortality and, indeed, it is difficult to improve
mortality rates in this day and age, said Dr. Sung, Chief of Gastroenterology
and Hepatology, Department of Medicine and Therapeutics, Prince
of Wales Hospital, Shatin, Hong Kong.
The abiding problem is how to identify high-risk patients. Led by
Alan Barkun, MD, of McGill University, Montreal, a largely Canadian
consensus group reviewed the literature to determine the barometers
of recurrent bleeding, including lab and clinical factors and the
presentation of the bleed (Ann Intern Med 2003;139:843-857).
This review identified the determinants of high risk as age (greater
than 70 years), hemodynamic shock, poor health status and comorbid
medical conditions. Low hemogobin, coagulopathy, hematemesis, melena
and fresh rectal bleeding were other parameters. "These patients
have a substantial risk of persistent or recurrent bleeding and
mortality and may require urgent management," Dr. Sung said.
As for non-endoscopic treatments that may buy a patient time, a
recent meta-analysis looked at 30 randomised trials of intravenous
H2 blockers in more than 3,000 patients with gastric and duodenal
ulcers (Levine et al, Aliment Pharmacol Ther 2002;16:1137-1142).
"Unfortunately, this drug was not proven to stop rebleeding
or to reduce the need for surgery and decrease mortality."
When proton pump inhibitors were introduced a decade or so ago,
there was hope that these drugs could stem ulcer bleeding. A UK
study of more than 1,000 randomized patients studied IV omeprazole
versus placebo and compared patients with melena and hematemesis
in both treatment arms (Daneshmend et al BMJ 1992;304:143-147).
"Endoscopy
was not routinely provided to all patients, but given at the discretion
of the endoscopist," he said. Disappointingly, this potent
acid suppressor was unable to achieve hematological control, conferring
no benefit in rebleeding, transfusion requirements, operations and
mortality.
A few years later, a study in India reported that oral omeprazole
stopped bleeding-a seemingly contradictory finding since no endoscopic
therapy was given even after endoscopic diagnosis. Although there
were reductions in rebleeding, surgery and transfusions, mortality
remained much the same. The main beneficiaries were patients with
a visible blood vessel or a blood clot. "We still don't know
the full potential of PPIs as pre-emptive therapy before endoscopy,
so further investigation would be worthwhile," Dr. Sung said.
Today, nobody would dispute that endoscopic therapy is the first
line of treatment for bleeding ulcers-especially epinephrine injection
or thermal or mechanical hemostasis. But what are the advantages
of using two therapies in combination? A Spanish study summarized
all the published literature on combining epinephrine injection
with another modality versus epinephrine alone [Calvet et al,
Gastroenterology 2004;126:441-50).
In 16 studies comprising more than 1,600 patients, combining epinephrine
with either mechanical or thermal hemostasis was able to reduce
recurrent bleeding and surgery and to decrease mortality by about
half. The study confirmed that it is best to secure hemostasis with
another modality and not try for just a transient hemostatic effect
with epinephrine alone.
At the Prince of Wales Hospital in Hong Kong, Dr. Sung's division
handles about 1,200 GI bleeds a year, of which 800 are due to ulcers.
His group collected data on 3,386 ulcer bleeders treated between
1995 and 1998, of whom one-third required endoscopic therapy. In
1.4 per cent of cases the index endoscopy was unable to stop bleeding.
"In most others, hemostasis was secure, but 8.2 per cent still
suffered recurrent bleeding, so roughly 10 per cent failed hemostasis
despite endoscopic therapy," he said.
To improve on such scenarios, physicians must identify high-risk
patients as defined above. When Dr. Sung and colleagues tested a
PPI combined with endoscopic therapy, they were surprised to find
reductions in recurrent bleeding, endoscopic retreatment, surgeries
and transfusion rates. There was also a substantial difference in
mortality, and the study was terminated prematurely for ethical
reasons.
Recently his group studied 54 patients whose ulcers were not actively
bleeding but were covered with a clot or showed a protuberant blood
vessel, and compared the effectiveness of a PPI alone for hemostasis
or endoscopy plus pharmacologic control. "We found that without
endoscopic treatment the number of patients with recurrent bleeding
actually builds up in the first 30 days." Nine of 54 had recurrent
bleeding, mostly patients with protuberant vessels and very few
with adherent clots. "If you remove a clot and see no protuberant
vessel, medical therapy alone should be sufficient."
A soon-to-be published Cochrane Review of about 3,000 patients in
21 randomised controlled trials has confirmed the benefits of PPIs
in reducing both recurrent bleeding (odds ratio: 0.46) and surgery
(OR: 0.59). Mortality, however, remains about the same. Why no improvement
in death rates? "It may be that even large studies that put
all the data together still do not have the power to detect a difference,"
Dr. Sung said. "Alternatively, the results might reflect poor
or deteriorating health status or terminal events, so that no matter
what we do, some patients die of other conditions."
Do recurrent cases warrant a second round of endoscopic therapy
or is surgery in order? Dr. Sung's colleague James Lau, MD, compared
endoscopy with surgery and concluded that surgery is still the gold
standard for managing bleeds (Lau et al, N Engl J Med 1999;340(10):751-56).
Three of 44 patients had rebleeds after surgery versus seven of
48 with endoscopy. "But the complication rate was significantly
higher with surgery, so we achieve better security of hemostasis
at the cost of a higher rate of surgical morbidity."
Dr. Sung stressed that the treatment of peptic ulcer bleeding should
start with classification of risk. For Forrest's I or II ulcers,
combining a PPI with endoscopic therapy is recommended. If the ulcer
is clean-based, an oral PPI alone should suffice. "If a patient
has recurrent bleeding, you can repeat endoscopic treatment or go
for surgery," he said.
|
