Page 20 - Gastrointestinal Bleeding (Xuất huyết tiêu hóa)

294 PART III Symptoms, Signs, and Biopsychosocial Issues

be started immediately and can be initiated on an outpatient basis patients should be treated with a daily PPI for 8 to 12 weeks and
when the patient has resumed a normal diet. In patients who are undergo repeat endoscopy to exclude underlying Barrett’s esoph-
found to have an Hp-induced ulcer, confirmation of the eradica- agus (see Chapter 47).
tion of Hp after treatment is recommended (see Chapter 52). Patients can sometimes present with mild UGI bleeding
from esophagitis not related to GERD but to infections (e.g.,
Aspirin, Other NSAIDs, and Antiplatelet Drugs Candida, HSV, CMV) or pill-induced esophagitis. Endoscopy
Ideally, patients with ulcer bleeding caused by aspirin or with biopsies and brushings is critical for making these diag-
another nonselective NSAID should stop the drug. If the noses and determining the appropriate pharmacologic therapy
patient is also positive for Hp, the organism should be eradi- (see Chapter 45).
cated with standard therapy (see Chapter 52). 162 In patients
with a history of ulcer bleeding who are positive for Hp and Ulcer Hemorrhage in Hospitalized Patients
need to continue taking low-dose aspirin (81 mg daily), eradi-
cation of Hp alone results in ulcer rebleeding rates similar to Hemorrhage from an ulcer or erosions in hospitalized patients
those associated with daily PPI therapy (if Hp is not eradi- typically falls into 2 categories. The classic cause is stress-related
cated). 163 By contrast, in patients with a history of ulcer bleed- mucosal injury (SRMI, or stress ulcers), characterized by diffuse
ing who are positive for Hp and need to continue full-dose bleeding from erosions and superficial ulcers. The second cate-
NSAID therapy, eradication of Hp alone without a PPI leads gory is inpatient ulcers, which are large, focal, chronic-appearing
to a significantly higher rebleeding rate than use of a daily PPI ulcers that are painless and present with severe UGI hemorrhage
in conjunction with the NSAID. In patients with ulcer bleed- manifested by hematochezia, melena, or bloody emesis. On
ing who do not have Hp infection but who need to continue emergency endoscopy, focal inpatient ulcers are often actively
daily aspirin, co-therapy with a daily PPI significantly reduces bleeding or demonstrate a visible vessel or adherent clot and are
the rebleeding rate compared with placebo. 164 Patients who marked by high rebleeding rates, despite combination endoscopic
require an antiplatelet medication such as clopidogrel and have therapy, and delayed healing on a high-dose PPI.
a history of ulcer bleeding will have less chance of recurrent SRMI occurs in the UGI tract of severely ill inpatients in an
bleeding if they take aspirin (81 mg) and a PPI daily compared ICU and is likely caused by a combination of decreased muco-
with taking clopidogrel alone. 165 sal protection and mucosal ischemia. SRMI usually occurs in the
Patients who require an NSAID after an ulcer bleed may be stomach but can also be seen in the duodenum, esophagus, and
considered for a selective COX-2 inhibitor. Selective COX-2 even rectum. Diffuse oozing is common, and patients have a poor
inhibitors cause fewer ulcers than nonselective NSAIDs but are prognosis and high rebleeding rate, often related to impaired
associated with a greater rate of cardiovascular complications. wound healing and multiple organ failure.
Because selective COX-2 inhibitors result in rebleeding rates Bleeding from SRMI is now uncommon, with a frequency of
similar to those associated with a nonselective NSAID and PPI approximately 1.5% of patients in an ICU. The 2 main risk fac-
co-therapy, their use may not be worth the increased cardiovas- tors are severe coagulopathy and mechanical ventilation for lon-
cular risk. 166 ger than 48 hours. 172 The frequency of clinically significant GI
bleeding with either or both of these risk factors is 3.7%, com-
Repeat Endoscopy to Confirm Gastric Ulcer Healing pared with 0.1% when neither risk factor is present. Other pro-
posed risk factors include a history of UGI bleeding, sepsis, an
Repeat EGD should be considered in patients with a gastric ICU admission longer than 7 days, occult GI bleeding for more
ulcer after 6 to 10 weeks of acid suppressive therapy to confirm than 5 days, and treatment with high-dose glucocorticoids.
healing of the ulcer and absence of malignancy (see Chapters ICU patients with risk factors for bleeding are the main tar-
53 and 54). In areas of the world where the population is at get groups for pharmacologic prevention of bleeding SRMI.
intermediate risk for gastric cancer, 2% to 4% of repeat upper Therapy with an H2RA has been shown to decrease the rate
endoscopies to confirm ulcer healing have been reported to of clinically significant bleeding in ICU patients at high risk of
disclose gastric cancer. 167-169 Some experts have suggested that SRMI. 173 One large multicenter study found that prophylac-
when the index endoscopy with biopsies is negative for malig- tic treatment with oral omeprazole or IV cimetidine results in
nancy and the ulcer appears benign endoscopically, a follow- similar bleeding rates, but that omeprazole is more effective than
up endoscopy is unnecessary. 170 A small retrospective study has cimetidine in maintaining the luminal gastric pH above 4. 174 A
found that when gastric cancer is detected on repeat endoscopy potential harmful effect of gastric acid suppression to prevent
to evaluate gastric ulcer healing, survival is no better than that stress ulcers is proliferation of bacteria in the stomach secondary
for patients who did not undergo the recommended follow-up to the increased gastric pH, and the associated risk of aspiration
endoscopy. 171 and ventilator-associated pneumonia; however, randomized trials
in which acid suppression (with an H2RA or antacids) and sucral-
fate (which does not lower gastric pH) were compared have not
Other Nonvariceal Causes shown convincingly that raising gastric pH increases the risk of
Esophagitis pneumonia. 175,176
Generally, if a patient with SRMI or an inpatient ulcer is sup-
Patients with severe erosive esophagitis can present with hema- ported hemodynamically and medically, the lesion will heal as the
temesis or melena. A multivariate analysis from a center in patient’s overall medical status improves. Because SRMI is diffuse,
France, in which 8% of all UGI bleeding was caused by erosive endoscopic therapy is generally not feasible. By contrast, focal
esophagitis, found that independent risk factors for bleeding inpatient ulcer hemorrhage often requires endoscopic hemostasis
esophagitis were grade 3 or 4 (moderate to severe) esophagitis by for severe hemorrhage (see Fig. 20.9); however, rebleeding rates
the Savary-Miller grading system (see Chapter 46), cirrhosis, a are higher and healing is slower than in patients in whom bleed-
poor performance status, and anticoagulant therapy. 171 A history ing starts before hospitalization (see Table 20.7). 177,178 A study
of heartburn was obtained in only 38% of patients. Severe bleed- in which epinephrine injection plus hemoclip placement was
ing from gastroesophageal reflux-induced esophagitis is treated compared with epinephrine injection plus MPEC in a cohort of
medically with a PPI (see Chapter 46). EGD is essential for diag- patients who had a high frequency of in-hospital ulcers found a
nosing severe erosive esophagitis, but endoscopic therapy gener- significantly lower rebleeding rate in the group that underwent
ally has no role unless a focal ulcer with a SRH is found. These injection and hemoclip placement. 133