Background: In the US alone, almost 700,000 endoscopic retrograde
cholangiopancreatographies (ERCPs) are performed each year, and the utilization of ERCP
over the last 10 years has increased due to the established therapeutic benefit of the
procedure. The most common complication of ERCP is the development of post-ERCP
pancreatitis (PEP). Depending on whether the ERCP is performed in average or high-risk
patients, the incidence of PEP ranges from 3-15%. The average Medicare reimbursement for
PEP is approximately $6,000 and the estimated annual cost burden of PEP is in excess of
$200 million. Anti-inflammatory prophylaxis with rectal indomethacin and pancreatic duct
stenting has been shown to reduce both the incidence of PEP and PEP severity. Yet, PEP
remains a common complication due to the suboptimal efficacy of these current
preventative modalities. The primary reason for the lack of progress in PEP prophylaxis
is the lack of novel approaches to target the underlying mechanisms of PEP.
In the initiation of acute pancreatitis, calcium is released by acinar cells, the main
parenchymal cell of the pancreas. Central to this pathway is the activation of the
heterodimeric calcium-dependent serine, threonine phosphatase calcineurin (Cn). In
addition to experimental evidence, recent clinical reports have demonstrated lower rates
of PEP in transplant patients taking Cn inhibitors. To gain an understanding of this
phenomenon, the investigators performed a search of the electronic medical records at the
University of Pittsburgh Medical Center, from 2005 to 2013, for patients who underwent
ERCP and found that tacrolimus users had close to a 50% reduction in PEP rates compared
to non-tacrolimus users (13.2% to 6.9%). A recent retrospective study showed similar
results. While these observations are subject to several confounders, including
co-morbidity and polypharmacy, the overall data provides both an experimental and
clinical premise for investigating the efficacy of Cn inhibitors in PEP.
In this trial, the investigators test the overarching hypothesis that tacrolimus,
administered as an oral loading bolus just prior to ERCP, will provide additive PEP
prophylaxis to the current standard of care, rectal indomethacin.
Hypothesis: H1: A combination of oral tacrolimus and rectal indomethacin is superior to
the use of rectal indomethacin alone, for the prevention of post-ERCP pancreatitis among
high-risk individuals.
H2: Oral tacrolimus is superior to placebo for the prevention of post-ERCP pancreatitis
among non-high-risk individuals.
Sample size justification: Based on the information from the earlier controlled trials,
the Investigators assume that the PEP incidence will be 7% in the Rectal Indomethacin
group and it will be reduced to 3% by the additional use of oral tacrolimus in
combination with rectal indomethacin A two-sided α=0.05, and a power of 0.8, will require
926 patients for the comparison between the combination of oral tacrolimus and rectal
indomethacin versus rectal indomethacin alone among high-risk individuals. Since 20% of
patients undergoing ERCP are expected to be at high risk based on the investigators'
prior experience, the investigators estimate that the total number of patients needed to
undergo ERCP is 4630. Adjusting for a 5% drop-out rate, the sample size needed will be
4874.
Recruitment and Consenting: Patients scheduled to undergo ERCP will be screened for
patient-based inclusion/exclusion criteria and will be consented to before the start of
ERCP.
Randomization: The randomization schedule is centrally generated at Johns Hopkins
University. Patients will be stratified by the participating center. Within each stratum
randomly varying block sizes of 30-50 will be used. Patients and treating physicians will
be blinded for the treatment allocation. This will be ensured by directly delivering the
list of randomly generated numbers to the investigational drug pharmacies.
Statistical Plan: The baseline characteristics of age, sex, comorbidity, American Society
of Anesthesiologists (ASA) score, ERCP indication, PEP risk factors, and the use of other
prophylactics (e.g. Intravenous Fluids) will be reported. Data will be presented in
percentages for categorical variables. Continuous variables will be presented as mean
with standard deviation (normal distribution) or median with interquartile range (skewed
distribution).
The modified intention-to-treat (ITT) principle will be applied to the primary analysis.
That is, all randomized patients will be analyzed according to the patients' original
treatment allocation, regardless of study protocol violations. The only patients excluded
from the analysis will be those in whom the duodenum was not reached and the papilla was
not manipulated (e.g., in case of upper gastrointestinal stenosis, aspiration risk,
restless patient) or follow-up could not be performed for 5 days. Because these patients
will not ultimately undergo an ERCP, there is no risk of PEP. A two-tailed P value of
less than 0.05 is considered to be statistically significant.