PHOSPHATIDYLETHANOL TESTING AS A DIAGNOSTIC TEST FOR OCCULT ALCOHOL MISUSE IN PATIENTS WITH ACUTE AND CHRONIC PANCREATITIS

Background
Germline genetic testing is recommended for younger patients with idiopathic pancreatitis, but there is no consensus recommendation for those over age 35. We aimed to analyze the results of genetic testing for pancreatic genetic mutations using a large dataset including all ages.

Methods
Individuals (ages 0-90) who underwent germline multigene testing for pancreatitis susceptibility genes (CASR, CFTR, CPA1, CTRC, PRSS1, SPINK1) through a large commercial laboratory between 2017 and 2022 for any cause were selected. Test results and information collected from requisition forms were evaluated. Multivariable logistic regression models were performed to identify factors associated with a positive pancreatitis panel result (defined as at least one pathogenic, likely pathogenic, and/or increased risk variant in a pancreatitis-related gene). Clinically significant results were defined as positive results in PRSS1, biallelic CFTR or SPINK1, or polygenic (i.e., CFTR and CASR).

Results
Overall, 2,468 subjects with primary indication of acute pancreatitis (AP; n=401), chronic pancreatitis (CP; n=631), pancreatic cancer (n=128), or other indications (n=1,308) completed panel testing. Among subjects with AP or CP (n=1,032), the most common variants were monoallelic CFTR (18.8%), monoallelic SPINK (6.9%), PRSS1 (4.5%), polygenic (2.0%), CTRC (1.6%), and biallelic CFTR (1.3%). The frequency of pathogenic variants varied according to age at the time of testing (Figure 1, 2). Variants in CASR contributed to polygenic results (with SPINK1) in 3 subjects. Among patients with pancreatic cancer, 10.2% had monoallelic variants in CFTR, but none had biallelic variants in CFTR.

Among subjects with AP or CP, the frequency of an abnormal result was significantly greater for those <35 versus ≥35 years of age (32.1% vs 24.5%, p=0.007). Similarly, the frequency of a clinically significant result was higher for those <35 vs ≥35 years of age (10.8% vs 5.4%, p=0.001). After adjusting for age, sex, race/ethnicity, primary indication for testing, and family history of pancreatitis, a positive family history of pancreatitis was associated with increased odds ratio (OR) of 8.59 (95% confidence interval (CI) 2.92-25.25) for a clinically significant panel result while each 5-year increase in age at test completion had lower odds (OR 0.89, 95% CI 0.83-0.95).

Discussion
The yield of germline genetic testing is highest in younger individuals with a positive family history of pancreatitis, which supports current recommendations for testing. However, a clinically significant result was found in approximately 5% of older adults. Therefore, we suggest that genetic counseling and germline testing should be considered for patients of all ages with acute or chronic pancreatitis.

Background: Development of organ failure (OF) within the first week of acute pancreatitis (AP) is characterized as early severe acute pancreatitis (ESAP) and has a mortality of up to 40%. The pathophysiology of OF in ESAP is not well understood. Our aim was to characterize cytokine storm syndrome (CSS) and secondary hemophagocytic lymphohistiocytosis (HLH) in AP.

Methodology: In this prospective observational cohort study, all consecutive patients with AP presenting within 14 days of onset of abdominal pain were included. Serum cytokines and chemokines were measured in all patients on day 3, 7, 14, 21 and 28 or till hospital discharge. The diagnosis of CSS was made in the presence of (i) elevated levels of serum IL-6 >160 pg/ml, (ii) presence of systemic inflammation such as systemic inflammatory response syndrome >48 hours and (iii) OF grade ≥2 persisting for ≥48 hours in the absence of sepsis. Secondary HLH was diagnosed as per 2004 HLH criteria or a H-score >169. CSS was correlated with clinical outcomes such as OF and mortality.

Results: From February 2021 till August 2022, 152 consecutive patients hospitalized for AP were screened and 98 patients were included in the study: 23 (23.5%) had mild, 24 (24.5%) moderate and 51 (52%) had severe AP as per revised Atlanta Classification. Of the 98 patients, 40 (40.8%) developed CSS with higher admission IL-6 levels [median 278 (IQR 207.5-450) pg/ml vs. 93.75 (43.1-154) pg/ml, p <0.01], modified Marshall OF score [Median 4 (IQR 3-5) vs. 0 (0-2), p <0.01] and higher in-hospital mortality [16 (40%) vs. 6 (10.3%), p=0.001] as compared to no CSS group (N=58). Serum TNF-a (r= 0.878, p=0.009), IL-6 (r=0.85, p=0.015), IL-8 (r=0.772, p=0.042), IL-10 (r=0.832, p=0.02), and IL-1b(r=0.884, p=0.008) showed statistically significant positive correlation with modified Marshall OF score on day 3 in CSS, while CCL-5 (r= 0.875, p=0.01) showed statistically significant negative correlation. Serial change in levels of IL-1b, IL-2, IL-6, IL8, IL-17 and IL-18 correlated with modified Marshalls OF score and mortality (Fig 1). Of the 51 patients with ESAP, 15 patients developed secondary HLH after a median of 9 (IQR 8-14) days from onset of AP HLH resolved on day 20 (15-27) of illness in patients who survived (n=8). Onset and resolution of HLH was in parallel with the onset and resolution of OF (Fig 2). No specific therapy for HLH was given [MOU1] and 6/15 (40%) died in HLH group.

Conclusion: Development of CSS in early phase of AP portended poor outcomes with higher OF and mortality. A proportion of patients with ESAP developed secondary HLH. These entities need to be recognized early to develop therapies targeting exaggerated dysregulated immune responses and improve outcomes in severe AP.

Introduction: Angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs) are used to treat hypertension; however, the use of ACE inhibitors has been suggested to cause acute pancreatitis (AP) in several case reports, and lack of epidemiological study to investigate this association. On the other hand, experimental studies have suggested a protective effect of ARBs on AP has been suggested, but clinical evidence is scarce. Hence, we aimed to explore whether ACE inhibitors and ARBs are associated with the incidence of AP.
Methods: We conducted a retrospective cohort study using the TriNetX database in the USA, comprising 9,929,954 new users of ACE inhibitors, ARBs, and dihydropyridine calcium channel blockers (dCCBs) between 2010 and 2021. Adult patients who were new users of ACE inhibitors and ARBs were matched 1:1 using propensity score matching (PSM) for demographic characters, BMI, nicotine dependence, alcohol dependence, comorbidities, HbA1C, baseline blood pressure, use of other cardiovascular medications and anti-diabetic medications with new users of dCCBs. The primary study outcome was incidences of AP. Risk ratios with 95% confidence intervals were calculated for all analyses.
Results: The first cohort included well-matched 304,083 patients in each group with new ACE inhibitor initiators and dCCB initiators after PSM, whereas the second cohort included a well-matched 2,660,521 patients in each group with new users of ARB initiators and dCCB initiators. During the follow-up (mean follow-up 5.8±1.2 yrs., ACE inhibitors vs. 4.9±1.1 yrs., dCCBs), 34,880 patients in the ACE inhibitors and 33,497 patients in the dCCBs cohort developed with AP after an exposure lag of 60 days (RR 1.04; 95% CI 1.02-1.05) (Table 1). In contrast (mean follow-up for ARBs was 5.3±1.0 yrs.), ARBs were not associated with an increased risk of AP compared to the new users of dCCBs (21,987 vs. 26,679; RR 0.82; 95% CI 0.81-0.84). However, there was no clear duration-response relation with AP. Hence, we performed a sensitivity analysis similar to the primary analysis by increasing the lag exposure to different timelines, such as at 6 months, 1 year, and 5 years. Outcomes were similar to the ones generated in the primary analysis.
Conclusion: ACE inhibitors were associated with an increased risk of AP compared with dCCBs, whereas ARB use was negatively associated with AP; hence, this supports that the use of ARBs has a potential protective association with AP. Therefore, clinicians should consider the possibility of ACE inhibitor-associated AP among patients currently taking or new users of ACE inhibitors and those presenting with the diagnosis of AP without a definite cause.

Background: HTG-AP may be associated with a higher risk of severe AP compared to other causes (PMID: 33087766). However, the mechanisms underlying the severity of HTG-AP are unknown. NEFAs are increased in both serum (PMID:12120200, 8501351) and necrotic collections (PMID: 25500204) in severe AP, and cause organ failure (PMID: 31917686). We therefore compared NEFAs, severity in patients with HTG-AP and AP from other causes.
Methods: Patients presenting to Mayo Clinic AZ Emergency Room with serum lipase > 3-fold upper limit of normal between August 2019 and June 2022 were included. Excluded were patients lacking admission triglycerides and those with active cancer. Patients were then sorted based on admission Triglyceride (TGL) levels with a cut off value of 500 mg/dL (HTG-AP) and etiology of AP based on chart review. These included HTG-AP, Alcoholic AP, Biliary AP, Idiopathic AP and other causes of AP including drugs, post ERCP pancreatitis. Admission serum TGL, NEFA (as µM by gas chromatography), age (years), BMI (kg/m²) were then compared between different etiologies using ordinary one-way ANOVA. ICU admission & sex were compared using a Chi-Square test. P values of <0.05 were regarded as significant.
Results: Among 466 patients, 324 fulfilled inclusion criteria. Of these, 27 patients had HTG-AP, 54 had alcoholic AP, 49 had Biliary AP, 139 had idiopathic AP and 55 had other causes of AP. Serum NEFA were significantly higher (p<0.004) in HTG-AP compared to alcoholic, biliary, idiopathic, other causes of AP (1172±751, 793±514. 748±395, 629±488, 667±303 respectively). Serum TGs in HTG AP were significantly higher (p<0.0001) than other etiologies (1633±4133, 198.2±120.2, 156.9±91.8,191.5±90.7, 176.9±99.2 respectively), while serum lipase in HTG-AP (1052±781 vs. 1381±1086U/L) was similar to others. The BMIs were similar (p>0.4) in all groups (29.1 ± 4.9, 27.6±5.6, 31.10±7.0, 28.0±6.3, 29.3 ±6.8). ICU admissions were also significantly higher in HTG-AP vs non-HTG-AP group (6/27 vs 27/299; p: 0.042) but not in Alcoholic AP vs non-alcoholic AP (4/60 vs 26/266; p:0.622), biliary vs non-biliary AP (9/54 vs 24/274; p:0.085) and other causes of AP (7/59 vs 26/262; p:0.63). ICU admissions were significantly lower in idiopathic AP compared to non-idiopathic AP (5/141 vs 26/175; p:0.0009). Alcoholic AP patients were younger than HTG AP (44.1 ± 16.3 vs. 53.6 ± 13.0 years p=0.045), though there were no differences in age of HTG-AP compared to other groups.
Conclusions: HTG-AP is uniquely associated with higher serum NEFA at the time of admission compared to other AP etiologies. This is associated with a higher rate of ICU admission compared to other etiologies. The higher serum NEFA in HTG-AP may be due to increased lipolysis of the elevated circulating TGs. This lipolytic generation of higher NEFAs in HTG AP may explain the greater severity noted in HTG-AP.

Background. Acute pancreatitis (AP) is one of the most common gastrointestinal diseases resulting in hospitalization. Recurrent acute pancreatitis (RAP) risk can be reduced by addressing the underlying etiology. Gallstone disease (GSD) is the most common etiology for acute pancreatitis (AP) and with risk associated with female sex, obesity and polygenic risk scores (PRS) (Lim, PMID: 34217876). Alcohol is the second most common etiology of AP with risk associated with men and a threshold risk at >4 drinks per day (e.g. very heavy drinking) with possible protection from AP with moderate drinking. We tested the hypothesis that alcohol reduces the risk of gallstones as a mechanism for potential protective effects from AP with moderate consumption.

Methods. We evaluated the UK Biobank (UKBB) cohort of 483189 subjects with complete genotype data. We computed the PRS for gallstones as published by Tanigawa Y et al. PLoS Genet (2022) using the PGS Catalog tools and categorized them by quintile for the entire cohort. Alcohol intake frequency was recategorized as never drinking to ~1/month, ~1/week, and >3days/week. GSD presence (GSD-pos) or absence (GSD-neg) was determined using ICD10-, ICD9-codes, death certificates, and self-reported interview data. Associations were compared using Chi squared test. Multiple factors were evaluated using logistic regression models

Results. In the UKBB GSD was noted in ~ 7.8% of subjects. We replicated the finding of Lim et al, where men were less likely to have GSD than women (OR: 0.582, CI 0.546, 0.622, p=1.25e-58), risk of GSD progressively increasing with BMI above 25 with BMI 25-30 62% and >30 266% higher, and ~linear increase in risk with PGS from lowest to highest quintiles (OR 2.890, p=0.0e+00) or deciles (OR 3.737, p=0.0e+00). The PRS effect was greater in women than men, and greater with higher BMI. When compared to non-drinkers+~1/m, the risk of GSD was less likely with drinking ~1day/week (OR: 0.772, CI: 0.752, 0.794; p=1.94e-74) and even less when drinking >3days/week (OR: 0.602; CI:0.586, 0.6179; p=1.140e-305), and GSD-pos subjects showed similar distribution of PRS across all alcohol intake frequency categories (p=0.316).

Conclusions. The risk of GSD is higher in women than men with progressive risk from higher BMI and higher gallstone PRS. The risk of GSD is reduced by 22.8% with ~1day/week of alcohol, and 39.8% with >3days/week of alcohol. This data suggests that the benefit of moderate alcohol consumption in reducing AP may be through reduced risk of biliary pancreatitis. However, after an initial AP attack the effects of any alcohol appears to increase risk of RAP and chronic pancreatitis. Further studies of GSD in alcohol-associated AP is warranted.

Background:
Hospital readmission rate is a key metric of patient care quality and represents a substantial burden to both the individual patient and the health care system. Necrotizing pancreatitis (NP) with its protracted course characterized by organ failure, need for intervention and ICU stay, malnutrition, and physical deconditioning, is associated with increased readmission rate. The aim of this study was to estimate the incidence and cause for 30 day unplanned readmission following index hospitalization for NP and to identify independent risk factors for readmission.
Methods:
Adult NP patients managed at our center between 2009-2022 were identified from a prospective database and categorized into 2 groups based on 30 day unplanned readmission following index hospitalization. Patients with no follow up, who died during index admission, or within 30 days of discharge were excluded. Baseline data on admission including demographics, ASA score, SIRS on admission and after 48 hours, multiorgan failure, NP interventions (endoscopic, percutaneous, or surgical), use of pain medications, nutritional intervention (enteral or TPN), and imaging characteristics were compared between the 2 groups. Multivariable analysis was completed to identify independent predictors of 30 day readmission; a p-value < 0.05 was considered significant.
Results:
Among 476 patients [males – 324 (68%), median age 51 years (IQR 38-63)] included, 192 (40%) had at least one unique readmission within 30 days of index discharge. There were 250 unique readmissions, with common etiologies for readmission being abdominal pain (36%), sepsis (24%), gastric outlet obstruction (4%), and feeding tube dysfunction (4%). On readmission, 125 patients (65%) required percutaneous/endoscopic/surgical intervention, 88 (46%) underwent necrosectomy, and 15 (8%) required ICU stay. Readmitted patients were at higher risk for mortality at 6 months (25% vs. 6%, P=0.03). Significant independent predictors of 30 day readmission on multivariable analysis were length of stay (LOS) ≥ 14 days at outside hospital (OSH) prior to transfer [OR 2.25 (1.15 – 4.38), P= 0.017], need for enteral nutrition on discharge [OR 1.76, (1.01 – 3.07), P= .046], and size of necrotic collection ≥ 6 cm [OR 2.12 (1.10 – 4.10), P= .025]. ASA II were 3 times more likely to be readmitted than ASA III as patients with ASA III were found to have longer length of stay (61.1 % v 30%, P-value <0.001).
Conclusions:
Readmission following NP is common (40%) and is associated with greater mortality at 6 months. Expedited transfer to tertiary center for timely care and intervention, assiduous follow up of other high risk patients (large collection and those who need enteral nutrition) could avoid readmissions and optimize outcomes.

Introduction
Recent moderate to high levels of alcohol intake is a modifiable risk factor for patients with acute and chronic pancreatitis. Blood ethanol levels may be low in symptomatic individuals since it is cleared rapidly and many patients will quit drinking in the days prior to their clinical presentation due to pancreatitis symptoms. The serum biomarker phosphatidylethanol (PEth) quantifies recent alcohol use and has high sensitivity and specificity for recent (approximately 30 days) alcohol intake – including binge drinking - and may be useful among patients without self-reported misuse. The objective of this study is to describe the frequency and characteristics of patients presenting with pancreatitis and elevated PEth to suggest recent alcohol misuse.

Methods
This is a retrospective cohort study including patients with a diagnosis of acute and/or chronic pancreatitis who underwent PEth testing in either the ambulatory or hospital setting between 1/2019 – 10/2022. PEth was measured in patients presenting with idiopathic pancreatitis or those with a remote history of alcohol misuse but self-reported abstinence or minimal use recently. Characteristics of patients with and without recent moderate to severe alcohol use (as defined by elevated PEth) were compared including presence of other behavioral factors for pancreatitis, overall comorbidity, and pancreas-related hospitalization within one year.

Results
During the study period, 99 patients presenting with acute (40), acute recurrent (53), and/or chronic (61) pancreatitis underwent PEth testing, 37 (37%) of whom had levels consistent with recent moderate to severe alcohol use. The majority (n=64) were checked in the ambulatory setting, while only 52 (53%) had a self-reported remote history of alcohol misuse. Patients with and without recent alcohol misuse (defined by PEth) were similar in terms of age, sex, BMI, overall comorbidity, and presence of diabetes mellitus (table). Patients with an elevated PEth level were more likely to have a history of alcohol misuse (37% vs. 3%, p<0.001), be an active smoker (54% vs. 8%, p<0.001), and be prescribed opioids (81% vs.59%, p=0.027). Additionally, these patients were more likely to have chronic pancreatitis at presentation (75% vs. 53%, p=0.026).

Conclusion
A substantial proportion of patients presenting with acute or chronic pancreatitis and no current alcohol misuse by self-report have an elevated serum PEth level, even those with no remote history of alcohol misuse; this suggests a high prevalence of occult alcohol misuse may be contributing to their clinical presentation. Since alcohol misuse is an important and modifiable risk factor for acute recurrent and chronic pancreatitis and PEth is known to be more specific than blood ethanol for recent alcohol use, PEth testing should be considered in the diagnostic evaluation.