Page 18 - Acute Pancreatitis (Viêm tụy cấp)

CHAPTER 58 Acute Pancreatitis 909

further attacks of pancreatitis. Alcoholic pancreatitis occurs more number of reports predicting moderately severe AP, which has
frequently in men approximately 40 years old. The first clinical been included in the revised Atlanta classification. 233 However, 58
episode usually occurs after 5 to 10 years of heavy alcohol con- despite a large body of literature, no perfect predictor is available
sumption. By contrast, biliary pancreatitis is more frequent in at the present time. Most predictors have a very high negative
women, and the first clinical episode is often after the age of 40 predictive value but not a useful PPV, and this is because a sig-
years. Recurrent attacks of AP suggest an alcohol etiology, but nificant proportion of the patients with AP do not develop mod-
unrecognized gallstones may cause recurrent pancreatitis. Among erately severe or severe disease. From 2 prospectively collected
patients with acute biliary pancreatitis discharged from the hos- cohorts, it was reported that the existing scoring systems seem to
pital without cholecystectomy, 30% to 50% develop recurrent have reached their maximal efficacy in predicting persistent organ
AP relatively soon after discharge (average time to recurrent pan- failure in AP. Sophisticated combinations of predictive rules are
creatitis, 108 days). 227 Thus removing the gallbladder in biliary more accurate but cumbersome to use, and therefore of limited
pancreatitis is imperative. clinical use. Our ability to predict the severity of AP cannot be
Laboratory tests may help distinguish between these 2 disor- expected to improve unless we develop new approaches. 234 A
ders. The specificity for gallstone pancreatitis of a serum ALT recent systematic review looking at predictors of persistent organ
concentration greater than 150 IU/L (≈3-fold elevation) is 96%; failure (severe AP) and infected pancreatic necrosis observed that
the PPV is 95%, but the sensitivity is only 48%. 205 The serum it is justifiable to use the blood urea nitrogen (BUN) level for
AST concentration is nearly as useful as the ALT, but the total prediction of persistent organ failure after 48 hours of admission
bilirubin and alkaline phosphatase concentrations are not as help- and procalcitonin for prediction of infected pancreatic necrosis
ful to distinguish gallstone pancreatitis from alcoholic and other in patients with confirmed pancreatic necrosis. 234 There was no
etiologies. There are differing reports as to whether a high serum predictor of persistent organ failure found that can be justifiably
lipase-to-amylase ratio can differentiate alcoholic from other used in clinical practice within 48 hours of admission. 235 What
causes of pancreatitis. 228,229 is really required is a system or a marker that has high PPV for
Conventional abdominal US should be performed in every moderately severe or severe AP. Along the same lines, the most
patient with a first attack of AP to search for gallstones in the recent AGA technical review on early management of AP found
gallbladder, common duct stones, or signs of extrahepatic bili- no studies using a predictive tool that improved clinical out-
ary tract obstruction. However, bile duct stones are frequently comes. 236 Hence it recommended using clinical judgment and a
missed by abdominal US, and most stones pass during the acute variety of predictive tools in clinical practice. It is for the same
attack. ERCP is limited to patients with severe AP due to gall- reason that this particular approach of predicting severity was not
stones with persistent bile duct obstruction and to those patients included in the meta-analysis in the same technical review. Thus
in whom the stone could not be removed during surgery. In most many of the predictors will be listed below briefly, with added
patients with biliary pancreatitis, common duct stones pass and information on some predictors that were widely studied. At the
no further evaluation is needed. Although the bile duct can be present time, most of these systems are more useful for research
imaged with an operative cholangiogram at the time of laparo- purposes and to compare different cohorts, rather than being use-
scopic cholecystectomy performed during the same admission, ful in directing clinical care.
this is not necessary in most patients. Recent guidelines and reviews recommended the following
Many clinicians prefer to evaluate the bile duct prior to sur- predictors at admission to be useful while considered together
gery. Because most stones that cause biliary pancreatitis pass, it is with clinical judgement: advanced age (>60 years), BMI, Charl-
not clear who should undergo evaluation. ERCP would be inap- son’s comorbidity index, pleural effusions or infiltrates on the
propriate in a patient with a moderate to low risk of choledo- admission chest radiograph, elevated hematocrit, elevated BUN
cholithiasis, when the risk of PEP is greater than the benefit of level, elevated serum creatinine level, and CRP >15 mg/dL at 48
ERCP (normal-sized bile duct and normal liver chemistry tests). hours. 1,211,237
However, if the bile duct is dilated and/or liver chemistry tests are The height of elevation of the serum amylase and lipase does
elevated, further evaluation prior to surgery may be reasonable. not correlate with severity. However, in pediatric AP, it was
Although EUS is an accurate method of detecting bile duct stones reported that a 7-fold elevation of lipase predicted severe dis-
and has been recommended for evaluating the bile duct prior to ease with a sensitivity, specificity, positive and negative predic-
cholecystectomy, it is rarely needed or used in this setting. 230 It tive values, and positive and negative likelihood ratios of 85%,
should be reserved for the patient who has findings suspicious for 56%, 46%, 89%, 1.939, and 0.27, respectively. 197 When AP is
a retained stone and cannot undergo MRCP. MRCP is as accu- superimposed on chronic pancreatitis, it is usually less severe than
rate, but because of its noninvasive approach, it is preferred if a AP without chronic pancreatitis. When superimposed on chronic
clinician has a lower suspicion that a common duct stone may be pancreatitis, weight loss, advanced age, and comorbidities pre-
present. If a common duct stone is found at surgery, it is either dict severity in a population-based study. 237 Radial EUS in acute
removed during the operation or endoscopically after surgery. 231 biliary pancreatitis showed a significant relationship between the
Laparoscopic exploration of the bile duct is as safe and effective as severity of AP with diffuse parenchymal edema, periparenchymal
postoperative ERCP in clearing stones from the common duct. 232 plastering, and/or diffuse retroperitoneal free fluid accumulation,
and peri-pancreatic edema, and also predicted mortality. 238 In a
PREDICTORS OF DISEASE SEVERITY large administrative database, transferred patients with AP have
more severe disease and higher overall mortality. Mortality is
According to the revised Atlanta classification there are 3 grades similar after adjusting for disease severity. Disease severity, insur-
of severity of AP: mild, moderate, and severe. Nearly 80% of the ance status, race, and age all influence the decision to transfer
patients have mild AP. Predicting the severity is very important patients with AP. 239
during the first 24 to 72 hours, mostly at admission and during A large number of clinical and laboratory predictors of severity
the first 24 hours. If such prediction suggests moderate or severe have been described in recent years. They include procalcitonin, 240
type of the disease, it may help communicate with the patient TNF-α, 241 thrombopoetin, 242 carboxypeptidase-B activation
about the course of the disease; triage them to intensive care or peptide, 243 polymorphonuclear elastase, PLA 2 , D-dimer in pedi-
step-up unit; and when specific interventions become available, atric AP, 244,245 higher urinary beta 2 microglobulin to saposin B
administer them early on. A variety of clinical features, labora- ratio, 246 hepcidin, 247 elevated levels of soluble B7-H2 (sB7-H2), 248
tory markers, and scoring systems have been described over the copeptin, 249 IL-6, 250 IL-17, IL-23, 251 melatonin, 252 resistin, 253
years to predict severe AP. There has been an extremely limited lower serum lipid concentrations, 254 mean platelet volume, 255