Digestive and Liver Disease - Volume 48 - Supplement 2

e202

Abstracts of the 22

National Congress of Digestive Diseases / Digestive and Liver Disease 48S2 (2016) e67–e231

outcome was 30-day mortality. We integrated the traditional risk

factors for mortality considered in the Rockall score with those

more recently recognized in the literature, e.g. inpatients bleeding,

endoscopic treatment failure, comorbidities severity (evaluated by

American Society of Anaesthesiologists physical status classification,

or ASA score), rebleeding and need for surgery. Statistics: the

performance yield of prognostic scores was assessed by comparing

the ROC curves

Results:

A total of 2,191 patients with ANV-UGIB were included

(mean age 69.2, 67.1% male). Comorbidities were present in 75.8%

and were judged as severe (ASA score 3-4) in 34.7% patients. At

admission, 7.2% of the patients had hemodynamic instability and

inpatient bleeding occurred in 410 (19.9%). Rebleeding and need for

surgery occurred respectively in 124 (5.6%) and 90 (4.1%) patients

with an overall mortality of 5.8%. in those patients, the native Rockall

score had a performance of 76% [AUC= 0.76 (0.73 to 0.80], while

when we integrated additional risk factors, the updated Rockall

provide a better performance [AUC= 0.85 (0.82 to 0.88) p <0.000,

Figure 1]. Compared the native score, the new Rockall has a better

sensitivity for death risk between 5 to 7 points (94.2% vs. 65.0%) and

an implemented sensitivity for score ≥ 8 points (81.7% vs. 21.7%,

p<0.000).

Conclusions:

In ANV-UGIB patients, different events could occur

during hospital stay, which in turn can increase the death risk.

Unfortunately, those are not considered during the initial clinical

triage of those patients. Our data show that when the traditional

Rockall score is implemented with new risk factors, the accuracy and

sensitivity are implemented, thus allowing a better identification of

patients with a higher mortality rate.

P.15.4

BLACK ESOPHAGUS: AN UNCOMMON CAUSE OF NON VARICEAL

UPPER GI BLEEDING

Massidda M.*

, Gaffuri N.

, Bettoni E.

, Genco C.

Istituto Clinico Humanitas Gavazzeni, Bergamo, Italy,

Istituto

Europeo Oncologico, Milano, Italy

Background and aim:

Black esophagus or acute esophageal necrosis

(AEN) is a rare cause of acute upper gastrointestinal bleeding

(AUGIB) characterized by circumferential black appearance of distal

esophageal mucosa that stops abruptly at gastroesophageal junction

(GEJ). It is diagnosed in elderly men with multiple comorbities and

younger adults with history of alcohol consumption. Etiology take

into account an initial ischemic damage and a topical injury that

can lead to diminished mucosal defense and compromission of the

intrinsic repair mechanisms. We report a case of AUGIB due to AEN.

Material and methods:

A 67-year-old man presented to the

emergency department with hematemesis after three days of fever,

right hypochondriac pain and vomiting. He reported moderate

recent alcohol consumption but no NSAIDs abuse. Laboratory

evaluation revealed conjugated hyperbilirubinemia with cholestasis,

hemoglobin 15.6 g/dL which decreased to 13 g/dl six hours later;

glycemia 467 mg/dl; BUN 241 mg/dL. On presentation, he was given

an iv bolus of proton pump inhibitor (PPI) 80 mg followed by an iv

PPI drip at 8 mg/h. A CT scan showed gallbladder, common bile duct

and intrahepatic gallstones with dilation of biliary tree.

Results:

An urgent upper endoscopy (EGD) revealed striking

diffuse circumferential black discoloration of the middle and

distal esophagus with abrupt interruption at the GEJ and no

active bleeding. Endoscopic biopsies were deferred and no further

endoscopic intervention was required. The patient was kept nil-

per-os with iv hydration, insulin and antibiotic therapy. Three days

after admission a second endoscopic look revealed white-yellow

circumferential exudates in the middle and distal esophagus. After

ERCP and cholecystectomy he progressively recovered without any

complication. He started soft diet at day 7 and was discharged in

good clinical conditions one week later.

Conclusions:

AEN is a rare cause (prevalence 0.001-0.2%) of AUGIB

with multifactorial etiology. Endoscopic appearance is diagnostic

and histologic confirmation is not warranted unless other etiologies

are suspected. Management consists of treatment of the underlying

diseases, fasting and high dose PPI. Antimicrobic therapy is indicated

only when infections are suspected. Mortality ranges from 13 to 35%

and esophageal perforation is reported up to 7% of cases. About

40% of patients develop progressive dysphagia due to esophageal

strictures. In conclusion, although AEN is a rare condition it should

be considered and mentioned on international guidelines as a

possible cause of AUGIB.

P.15.5

GUIDE WIRE ASSISTED CANNULATION OF MINOR PAPILLA IN

PANCREAS DIVISUM: OUR EXPERIENCE

Sica M.*, Mutignani M., Forti E., Pugliese F., Tringali A., Manta R.

Surgical Digestive Diagnostic and Interventional Endoscopy, “Niguarda

Ca’ Granda Hospital”, Milano, Italy

Background and aim:

The minor papilla cannulation is a challenging

procedure with an high rate failure in literature (20%)*. We describe

our experience of endotherapy in patients with pancreas divisum

affected by symptomatic pancreatic diseases.

Material and methods:

From April 2012 to October 2013 14 patients

with pancreas divisum affected by symptomatic diseases underwent

ERCP using a minor papilla approach. We retrospectively evaluated

technical results at first attempt and early complications as well.

Results:

In patients with a diagnosis of pancreas divisum (11/14)

we approroached directly to the minor papilla with a double lume

sphincterotome (Mini-Tome, Cook) and a 0.018 fr hydrophilic guide

wire without contrast injection (WIRE-GUIDE CANNULATION), then

we performed sphincterotomy and plastic stent placement (calibre

5, 8.5 or 10 Fr).

The minor papilla cannulation was achieved in 13 of 14 patients

(92.8%) at first attempt, in one case, after unsuccessful pre cut, we

repeated ERCP the next day and we got a successful cannulation

after secretin’s injection for a better visualization of duct’s opening.

Endoscopic minor papillotomy (EMP) was performed in 13 of 14

patients using a sphincterotome (Mini-Tome, Cook). Pancreatic stent

was placed in 92.8% of cases. In 3 cases we need to remove the stent

repeating the endoscopy, in the others we observed a spontaneuos

migration.