Statins and Post-ERCP Acute Pancreatitis (Stark Project)

Introduction:

Acute Pancreatitis (AP) is the most frequent complication after ERCP; its incidence is around 3.5-4.5%. Post-ERCP AP (PEAP) is defined as a clinical pancreatitis (new or worsened abdominal pain) with amylase/lipase at least three times the upper limit of normal at more than 24 hours after the procedure, requiring hospital admission or a prolongation of planned admission. It is resulting in increased morbidity of patients undergoing this technique, and an increased in health spending, due to prolonged hospital days and patient needs. In this context, different strategies have been investigated to prevent PEAP, for example rectal NSAIDs (indomethacin or diclofenac 100 mg just before or after procedure) and pancreatic stents.

On the other hand, statins are widely used drugs aimed to lower cholesterol blood levels (the frequency of statin use, may be as high as 50% of men and 36% of women between the ages of 65 years and 74 years in USA). The relationship between statins and AP has been controversial. In the past, it was believed that statins were associated to an increased risk of AP. In Data Sheet of different International Drug Agencies, we can find their mainly indications (hypercholesterolemia, mixed dyslipidaemias, hereditary hypercholesterolemia) and commons adverse effects with reference to acute pancreatitis (myopathy, rhabdomyolysis, hepatitis/jaundice, acute pancreatitis…). Besides, some studies, most of them case series or case reports, have associated the use of different statins with drugs induced AP.

However, recent evidence suggests the opposite: Statins could reduce the risk of AP. In a retrospective longitudinal cohort study on data from a big health database from California (USA), with a big population of three millions nine thousand people, the incidence of AP among subjects who used simvastatin was significantly lower than who didn´t use it (0,80 (0,80-0,81)/100.000 consumers/day versus 1,28 (1,27-1,28) /100.000 no consumers/day; crude incidence rate ratio for risk of AP with simvastatin use 0.626 (95% confidence interval (CI) 0.588-0.668, p<0.0001). In this study patients who received simvastatin were more likely to have gallstone-related disorders, alcohol dependence or hyper-triglyceridemia compared with the reference population; nevertheless, risk of AP was significantly reduced with simvastatin use. In multivariate analysis, simvastatin was independently associated with reduced risk of pancreatitis, adjusted RR 0.29 (95%CI 0.27-0.31) after adjusting for age, gender, race/ethnicity, gallstone disorders, alcohol dependence, smoking and hypertriglyceridemia. Similar results were noted with atorvastatin. In a Meta-analysis of 21 randomized controlled trials investigating effects of lipid-modifying therapies on cardiovascular events, was assessed the relationship between lipid modifying therapies and risk of AP. In this study it was concluded that the use of statin therapies was associated with a lower risk of AP, RR 0.79 (95%CI 0.65-0.95; p=0.01). In a case-control study from Denmark, it was examined if statins use is associated with increase risk of acute pancreatitis. 2576 first-time admitted cases of AP were controlled with 25 817 gender-matched controls from the general population. Prescriptions for statins prior to admission with acute pancreatitis or index date among controls were retrieved from prescription databases. Its findings speak against a strong causative effect of statins on the risk of acute pancreatitis, and speculated about it could indicate a mild protective effect.

Some studies even speculate about an improvement in the severity of AP among patients with prior use of statins. In a prospective cohort study from Croatia, which included patients with acute pancreatitis divided into two groups according to statin use prior to hospitalization, severe disease was more common in the no-statin group than in statin group (20.6% vs. 8.7% p=0.001), and severity markers were also higher in the no-statin group.

It isn´t clear what is the mechanism by which statins might have a beneficial effect on acute pancreatitis. Basic studies suggest that statins could act on a specific anti-inflammatory pathway, inhibited them, as hypothesize experimental models. Statin effect that may be relevant to AP include blockade of the interleukin 6 mediated Janus kinase/signal transducer pathway and up regulation of the unfolded protein response helpful in modulating the endoplasmic reticulum stress response.

Based on the above evidence, the study hypothesis is that statins consumption is associated to a decreased incidence of PEAP. A prospective cohort study is proposed, focused on patients undergoing ERCP, with the aim to compare the incidence of PEAP among patients consuming statins and those who do not.

Material and Methods:

Hypothesis: Statins consumption decreases incidence of PEAP. I. Main aim: Compare the incidence of PEAP among patients consuming statins and those who do not.

II. Secondary aims:

• Investigate the effect of other drugs, especially acetylsalicylic acid (ASA), in the incidence of PEAP.
• Study the statin mechanism to decrease the incidence of AP.
• Study the relationship between use of statins and the severity of AP.

Design Study: International multicentric cohort study (observational, analytical, prospective), focused on patients undergoing ERCP, with the aim to compare the incidence of PEAP among patients consuming statins (any statin) and those who do not.

Data acquisition:

Data were prospectively acquired by researchers from the participating centers. After informed consent, the patients were interviewed before and immediately after ERCP. According to local clinical practice, most of the patients were admitted for 24 hours after the procedure to monitor for possible post-ERCP complications, including PEAP. Patients discharged earlier were instructed to return in case of concerning symptoms (persistent abdominal pain, nausea, vomiting, fever, gastrointestinal bleeding etc.). Names and identification data of the patients were not recorded on the electronic case report form in order to ensure confidentiality in accordance with data protection laws. Data monitoring was performed by one of the collaborators. Demographic and analytical data, comorbidities, treatments received (more specifically statins and ASA), pancreatic diseases, procedure results and post-procedural complications were collected.

Sample calculation:

The sample size calculation was based on a presumed 5% incidence of PEAP among non-statin users, an expected 1:3 ratio of statin users to non-users, and a 70% decrease of PEAP rate among statin users. The alpha-error was set to 0.05 and beta-error was set to 0.20. A sample of 1016 patients was accordingly calculated.

Statistical analysis:

The results are expressed as the mean (standard deviation: SD), median (interquartile range: IQR), or n (%). Normality was assessed by means of the Shapiro-Wilk test. For bivariate analysis, quantitative variables were compared with qualitative variables by means of the Student t-test and Mann-Whitney U-test for 2 categories or with ANOVA/Kruskal-Wallis tests for more than 2 categories. Qualitative variables were compared with the chi-square test or Fisher's exact test if needed. Multivariate analysis was performed by means of binary logistic regression. Incidence, odds ratio (OR), and adjusted OR (aOR) were used as measures of the frequency and strength of association, and 95% confidence intervals (CIs) were calculated for OR and aOR. Both OR (95% CI) and aOR (95% CI) were calculated by means of binary logistic regression (bivariate and multivariate analysis, respectively). The multivariate model included those variables that reached statistical significance in the bivariate analysis (p < 0.05), those variables considered as definite risk factors of PEAP according to the 2014 ESGE guidelines [1], and the prophylactic measures such as periprocedural rectal administration of diclofenac or indomethacin or placement of a pancreatic stent. All statistical tests were 2-tailed, and p-values of less than 0.05 were considered statistically significant.