Su1072
Medication Exposures Associated With Increased Short-Term Mortality of
Patients With Clostridium difficile- Associated Disease (CDAD)
Rohit Das, Paul Feuerstadt, Erik Rahimi, Lawrence J. Brandt
Background & Goal: Patients commonly are exposed to various medications during the
AGA Abstracts
incubation period and time of diagnosis of C. difficile infection. Our hypothesis is that
exposure to specific medications during C. difficile infection is predictive of different outcomes
in pts with CDAD than in patients without CDAD. Methods: We retrospectively identified
consecutive pts who had a C. difficile toxin assay obtained at Montefiore Medical Center.
Our case group (Grp 1) included all pts with a positive toxin assay. Our control group (Grp
2) consisted of age-, sex- and Charlson co-morbidity score-matched controls with diarrhea
and negative toxin assay(s). We recorded various demographic factors and medication
exposures during the incubation period and at the time of toxin assay for C. difficile. Acid
suppressive medications, antibiotics, anti-fungal and anti-viral agents, NSAIDS and opiates
(82 total medications) were considered. Our primary end-point was 30-day mortality and
secondary end-points included ICU stay, surgical intervention and whether the patient
remained hospitalized 30 days after toxin assay. Outcome analyses were performed for each
medication considering each end-point and statistical analysis completed using SPSS (16.0)
software. Results: Grps 1 and 2 consisted of 1096 and 1065 pts, respectively. Grp 1 had a
higher mortality rate (15.3% vs. 10.8%, p < 0.01) and rate of pts remaining hospitalized
for >30 days after toxin assay (11.9% vs. 6.7%, p < 0.001). There were no significant
differences between groups 1 and 2 of rates of ICU stay and surgical treatment. After
considering age, Charlson co-morbidity score and all medications that independently pre-
dicted 30-day mortality with statistical significance, significant multivariate predictors
included: Grp 1: age (p<0.001), acetaminophen (p= 0.002), aztreonam (p= 0.002), piperacil-
lin/tazobactam (p=0.001), gentamycin (p<0.02), oxycodone/acetaminophen (p<0.05) and
fentanyl (p=0.02); Grp 2: Age (p<0.001), Charlson score (p<0.001), vancomycin IV (P=
0.001), cefipime (p<0.01) and fentanyl (0.013). When a similar analysis was performed
using ≥30-day hospitalization after toxin assay as its end-point, significant multivariate
predictors included: Grp 1: Acetaminophen (p=0.002), oxycodone/acetaminophen (p<0.05),
fentanyl (p<0.01), PPI use (p<0.02) and vancomycin IV (p<0.01). Grp 2: Fentanyl use (p=
0.02) and vancomycin IV (p<0.001). Conclusions: Patients with CDAD were hospitalized
for > 30-days more frequently and had higher mortality rates compared with non-CDAD
patients who had diarrhea. Acetaminophen, certain antibiotics and fentanyl administration
predicted short-term mortality in patients with CDAD compared with patients who had
symptomatic non-CDAD diarrhea. The use of acetaminophen and proton pump inhibitors
are unique predictors for prolonged hospitalization in patients with CDAD.
Su1073
Characteristics of Patients Who Switch From Twice Daily (BID) Proton Pump
Inhibitors (PPI) to Once-Daily (QD) Dexlansoprazole or QD Other PPIs
Onur Baser, Reema Mody, Robert Morlock, Lin Xie, Erdem Baser
Introduction: Approximately 12% to 20% of patients with gastroesophageal reflux disease
(GERD) require BID PPI therapy to control their symptoms. Literature suggests that in the
majority of patients on BID PPI therapy, step down therapy can be successful. The objective
of this study was to understand characteristics of GERD patients who switched from BID
PPIs to QD dual delayed release dexlansoprazole (DEX) or other single release PPIs. Methods:
Using data from a large U.S. health plan, adult GERD patients (ICD-9 codes 530.1, 530.10-
530.12, 530.81, or 787.1) having ≥1 PPI Rx claim (index date) between February 2009
to September 2009 with continuous eligibility for 12 months before and 6 months after the
index date were identified. Patients were classified as those who switched from pre-index
BID PPI to either QD DEX or other QD PPI at index date. The 2 groups were compared
using univariate and multivariate analysis on various clinical, demographic and treatment
characteristics. Results: Among 185,506 GERD patients identified, 17.6% (n=32,718) were
classified as BID PPI users during the pre index period. Approximately 2.0% (n=649) of
BID PPI users switched to QD DEX and 9.3% (n=3052) switched to QD other PPIs. Patients
who switched to DEX vs. other QD PPIs were younger (50.8 vs. 52.1 yrs, p<0.001), more
likely to be female (66.6 vs. 61.7%, p<0.001) and reside in the southern part of the United
States (59.8 vs. 53.2%, p=0.002). A higher proportion of DEX switchers had higher GERD
severity, as indicated by diagnosis of barrett's esophagus (10.2 vs. 7.0%, p<0.005) and
strictures and stenosis of esophagus (6.9 vs. 4.3%, p<0.003), more pre-index PPI fills (6.6
vs. 5.4, p<0.001) and higher pre-index daily consumption of medication (1.69 vs. 1.59,
p<0.001) compared to those switched to other PPI. A higher proportion of DEX switchers
had at least one GERD-related office (93.5 vs. 86.4%, p <0.001) and outpatient visit (54.7
vs. 40.2%, p<0.001) but lower proportion of GERD-related inpatient admissions (4.2 vs.
7.6%, p<0.002) in the pre-index period compared to those who switched to other PPIs. In
the post-index period, after risk adjustment, DEX switchers were less likely to discontinue
their treatment (49.8 vs. 56.6%, p=0.018), had longer length of therapy (119.87 vs. 111.47
days, p=0.021) compared to those who switched to other PPIs. There were no significant
differences in overall and GERD-related total healthcare resource costs, except for GERD-
related office visits and total outpatient visit costs where DEX switchers had higher costs
compared to those switched to other PPIs. Conclusion: Patients who switched from BID
PPIs to QD DEX are less likely to discontinue their treatment, have longer length of therapy
and have similar overall and GERD-related total health care resource costs compared to
patients who switched from BID PPIs to other QD PPIs.
AGA Abstracts
Su1074
The Use of Alosetron on Demand is Efficacious in Patients Experiencing
Constipation on Standard Continuous Dosing
Charles W. Randall, Franz Zurita, Christopher A. Fincke, Carlo M. Taboada, Russell D.
Havranek, David L. Stump, Gary S. Gossen
INTRODUCTION Since reintroduction in 2002, alosetron has been the only FDA approved
drug for the treatment of female patients with IBS-D. The most common adverse event is
constipation (9% on 0.5mg BID). This may occur when the drug has effectively relieved
diarrhea and urgency. The purpose of this study was to determine if alosetron can be used
as needed to control diarrhea. METHODS Female patients with IBS-D diagnosed per Rome
II or III guidelines taking alosetron were eligible to participate if they had adequate symptom
relief but experienced constipation defined as no defecation for greater than 48 hours.
Alosetron at 0.5mg BID was initiated on each patient at onset of diarrhea and discontinued
2 days after bowel movements normalized. If constipation occurred within first 48 hours,
the dose was decreased to 0.5mg per day at onset of next cycle. Should constipation be
persistent the patients could decrease dosing interval to every other day. There were no
restrictions as to how often alosetron could be used. The study design was prospective and
open label. The primary objectives were normalization of bowel movements as defined as
a Bristol Stool Scale score from 3-5 and absence of constipation. Patients were followed
regularly for 6 months duration. The primary objective was to determine if on demand
therapy restored normal defecation and reduced constipation. RESULTS 12 patients satisfied
criteria to begin the study. All completed the required 6 month evaluation. 8 patients
responded to on demand dosing at 0.5mg BID, 4 patients required only 0.5mg daily to control
symptoms, and none decreased the interval to every other day. No patients experienced
constipation or other adverse events. CONCLUSIONS 1. On demand therapy is a safe and
effective means of treating female patients with IBS-D intolerant to continuous therapy. 2.
Doses may be adjusted based upon patient response. 3. No constipation or adverse events
were observed.
Su1075
Laparoscopic Sleeve Gastrectomy (LSG) for Obesity: Consequences on
Gastroesophageal Reflux Disease (GERD) Symptoms, Characteristics of Reflux
Events and Esophageal Motility
Kafia Belhocine, Eric Letessier, Emmanuel Coron, Guillaume Boulanger, Jean Paul
Galmiche, Stanislas Bruley des Varannes
Introduction: GERD symptoms and esophagitis are significantly associated with obesity. LSG
is increasingly used to treat morbid obesity. However the consequences of this surgical
procedure on the physiological antireflux barrier are poorly known. The aims of this study
were to compare the prevalence of reflux symptoms as well as several functional esophageal
parameters (acid exposure and esophageal motility) before and after LSG. Methods: Thirty-
four consecutive obese patients (29 women, BMI: 46±7 kg/sqm, age: 47±11 yrs) were
enrolled in this prospective study. All of them were eligible for LSG and 8 had esophagitis/
Barrett's esophagus at preoperative upper GI endoscopy. A systematic work-up included a
standardized questionnaire for GERD symptoms (typical and atypical symptoms), a 24h-
esophageal pH-monitoring and a high resolution manometry (36 sensors - Sierra Inc, Los
Angeles, CA, USA) was performed prior to LSG and during post-operative follow-up (5 to
20 months - mean 9 months). Results: A diagnosis of GERD was established (abnormal acid
exposure and/or mucosal breaks at preoperative endoscopy) in 14 (41%) and 24 (71%)
patients before and after LSG respectively (p<0.01). No preoperative characteristic was
significantly associated to an abnormal postoperative esophageal acid exposure. Conclusion:
This study 1) confirms the high prevalence of GERD in obese patients, 2) shows that LSG
albeit reducing weight further aggravates reflux in some patients. The mechanisms of this
aggravation deserve further studies.
Su1076
Transoral Incisionless Fundoplication With EsophyX for Gastro-Esophageal
Reflux Disease: Long-Term Results and Pre-Procedure Findings Affecting
Outcomes
Pier Alberto Testoni, Cristian Vailati, Sabrina G. Testoni, Maura Corsetti
Background: transoral incisionless fundoplication (TIF) with the EsophyX™ device is
reported to be effective for creating a continent gastro-esophageal valve and for good func-
tional results, as measured by pH-Impedance study in patients with gastro-esophageal reflux
disease (GERD). Aim: to assess the long-term effect of TIF in patients with symptomatic
GERD. Methods: TIF fundoplication was done in 36 consecutive patients. All were studied
with GERD-HRQL and GERD-QUAL questionnaires, upper GI endoscopy, esophageal mano-
metry and 24h pH-impedance before, 6, 12 and 24 months after TIF. Results: In all, 29
patients completed a 6 months' follow up: 17 (58.6%) completely stopped proton pump
inhibitor (PPI) therapy, 5 (17.3%) more than halved it, and 7 (24.1%) patients continued
with the same dose as before the procedure. There were 22 patients with a complete 24-
month follow up: 9 (40.9%) completely stopped PPI therapy, 5 (22.7%) more than halved
it, and 8 (36.4%) were taking the same dose as before the procedure. The effectiveness of
TIF procedure, though good, did not reach the statistical significance (p=0.2). Hiatal hernia
S-398