Page 23 - Gastrointestinal Bleeding (Xuất huyết tiêu hóa)

CHAPTER 20 Gastrointestinal Bleeding 297

Patients with an acute UGIB and a history of an aortic aneu-
rysm repair should undergo urgent CT with IV contrast or MR 20
angiography first. CT or MRI may show inflammation around
the graft and may demonstrate the fistula. If these are not
diagnostic, push enteroscopy should be considered to evaluate
the third portion of the duodenum for compression, blood, or
graft material, as well as to exclude other bleeding sources. A
vascular surgery consultation should also be obtained. Surgical
treatment is required to remove the infected graft. Therapeu-
tic endoscopy plays no role in the management of bleeding
from an aortoenteric fistula (see Chapter 38).
Varices

Variceal hemorrhage is an important cause of UGI bleeding
and is discussed in more detail in Chapter 92. Esophageal vari-
ceal bleeding related to portal hypertension is the second most
common cause of severe UGI bleeding (after PUD). The acute
mortality rate with each bleed is approximately 30%, and the
long-term survival rate is less than 40% after 1 year with medi-
cal management alone. 211 Despite advances in medical therapy,
endoscopic hemostasis, and angiographic procedures and TIPS,
overall long-term survival rates have not improved for patients
Fig. 20.17 Endoscopic appearance of the ampulla of Vater and hemo- with variceal bleeding. Survival in nontransplanted patients with
bilia. Note fresh red blood on the right side exuding from the ampulla variceal bleeding is heavily influenced by the severity of under-
of a patient who earlier that day had undergone a percutaneous liver lying liver disease, with poorer survival rates for patients with
biopsy. higher MELD scores or Child-Pugh class C cirrhosis than for
those with Child-Pugh class A or B cirrhosis (see Chapters 74,
pancreatic pathology if previously unsuspected. Endoscopy with 92, and 97). LT can improve survival in selected patients.
a side-viewing duodenoscope reveals blood coming out of the Bleeding gastric varices are a difficult therapeutic problem
ampulla. Management of severe hemorrhage is usually with angi- because in contrast to bleeding esophageal varices, most available
ographic embolization or surgery. nonsurgical treatments are ineffective, except when isolated gas-
tric varices are found without accompanying esophageal varices,
Postsphincterotomy Bleeding as occurs with splenic vein thrombosis and often in association
with pancreatitis or pancreatic cancer. The diagnosis of splenic
Bleeding following endoscopic sphincterotomy occurs in approx- vein thrombosis can be made with Doppler US, MRI, or routine
imately 2% of patients (see Chapter 42). 206 Potential risk factors angiography. Bleeding from gastric varices caused by splenic vein
include coagulopathy, use of anticoagulants, portal hyperten- thrombosis is treated by splenectomy. Focal gastric varices with
sion, renal failure, and the type and length of sphincterotomy. bleeding can be treated with injection of cyanoacrylate glue or
Successful hemostasis of postsphincterotomy bleeding is usually radiologic procedures such as balloon-occluded retrograde trans-
achieved with endoscopic methods such as injection of epineph- venous obliteration (see Chapter 92).
rine, hemoclips, or MPEC (see Chapter 42).
Medical Management of Acute Variceal Bleeding
Aortoenteric Fistula Somatostatin and its long-acting analog, octreotide, cause
selective splanchnic vasoconstriction and lower portal pres-
Bleeding from an aortoenteric fistula is usually acute and massive, sure without causing the cardiac complications seen with
with a high mortality rate. 207 A primary aortoenteric fistula is a vasopressin (even in combination with nitroglycerin). Studies
communication between the native abdominal aorta (usually an ath- have shown mixed results as to whether somatostatin is more
erosclerotic abdominal aortic aneurysm) and, most commonly, the effective than placebo in managing variceal bleeding, but it
third portion of the duodenum. 208 Often, a self-limited herald bleed seems to be at least as effective as vasopressin and is much
occurs hours to months before a more severe exsanguinating bleed. safer. A meta-analysis has shown that vasoactive drugs (e.g.,
Occasionally, the diagnosis of an aortoenteric fistula is suspected octreotide, somatostatin, terlipressin [a long-acting vasopres-
by a history of an abdominal aortic aneurysm or by palpation of a sin analog]) are as effective as sclerotherapy for controlling
21
pulsatile abdominal mass. The diagnosis can be difficult to make on variceal bleeding and cause fewer adverse events. No studies
endoscopy in the absence of active bleeding. Demonstration of an have shown a survival benefit to vasopressin or somatostatin
aortic aneurysm on abdominal CT or MRI (with IV contrast) sug- in patients with variceal bleeding. Given the potential abil-
gests the diagnosis of a fistula. Secondary aortoenteric fistulas are ity of octreotide to control acute variceal hemorrhage, its low
58
more common and usually occur between the small intestine and an toxicity, and its availability in the USA, octreotide has been
infected abdominal aortic surgical graft. The fistula typically occurs the pharmacologic drug of choice as an adjunct to endoscopic
between the third portion of the duodenum and the proximal end of therapy for the treatment of variceal hemorrhage. The dose
the graft but may occur elsewhere in the GI tract. The fistula usually of octreotide for acute variceal hemorrhage is a 50 μg bolus
forms between 3 and 5 years after graft placement. Patients often followed by a continuous IV infusion of 50 μg/hr for up to 5
experience a herald bleed that is mild and self-limited, and occa- days.
sionally intermittent, before massive bleeding occurs. 209 A second- Patients with a prolonged PT that does not correct with fresh
ary fistula can also occur between the third part of the duodenum frozen plasma may benefit from infusion of human recombinant
and an endovascular stent, in which case the fistula may be caused factor VIIa, although prolongation of the PT does not corre-
by pressure from the stent against the duodenum, infection of the late with bleeding risk (see Chapters 92 and 94). In one uncon-
stent, or possibly expansion of the native aneurysm. 210 trolled trial, a single 80 μg/kg dose of recombinant factor VIIa