Article, Gastroenterology

Assessing the CT findings and clinical course of ED patients with first-time versus recurrent acute pancreatitis

a b s t r a c t

Study objectives: The primary objective of this study was to compare Emergency Department patients with first- time versus recurrent acute pancreatitis.

Methods: This study was a retrospective chart review of patients with a diagnosis of acute pancreatitis who pre- sented to a single academic urban emergency department from 2012 to 2016. Criteria for inclusion were clinical symptoms of pancreatitis, age greater than or equal to 18 years, ED diagnosis of acute pancreatitis, and an abdom- inal CT scan within 24 h of triage. Exclusion criteria were traumatic mechanism and pregnancy. Charts were reviewed by a minimum of two trained abstractors using structured data collection sheets and discrepancies were resolved by a third abstractor. Patients with first time acute pancreatitis versus recurrent acute pancreatitis were compared to determine differences in characteristics, management and disposition.

Results: 250 patients were included in the study. Of these, 165 patients had first-time acute pancreatitis and 85 patients had recurrent acute pancreatitis. Demographics, vital signs and initial lab values were the same in both groups. Patients with recurrent acute pancreatitis were more likely to have significant findings on CT (Modified CT Severity Index, 2.09 versus 1.43, p b 0.05), more likely to require IV opiates (96% versus 75%, p b 0.001) and less likely to need ICU admission (8% versus 19%, p = 0.03).

Conclusion: ED patients with recurrent acute pancreatitis demonstrated more significant findings on CT compared to patients with first-time acute pancreatitis but were less likely to require ICU admission.

(C) 2018



Acute pancreatitis has an incidence of 5 to 35 per 100,000 people in the US and the incidence is rising [1,2]. Diagnosis is made through a combination of clinical symptoms, blood tests and imaging. Due to the severity of the pain and the inability to tolerate food or liquids, acute pancreatitis is commonly treated in the ED [3]. 17% of patients with acute pancreatitis develop recurrent acute pancreatitis and 8% of pa- tients with Recurrent episodes develop Chronic pancreatitis [4]. Severe acute pancreatitis is associated with a mortality rate of 16.3% [5].


While the cornerstones of diagnosis of acute pancreatitis are the clin- ical presentation and elevated Pancreatic enzymes, CT scan is commonly

* Corresponding author at: 2120 L Street NW Suite 450, Washington, DC 20037, United States of America.

E-mail address: [email protected] (A.C. Meltzer).

used to confirm the diagnosis and to estimate the severity of the disease [6]. Current risk stratification tools do not consider whether acute pancre- atitis is recurrent or first-time as a marker of severity. Greater awareness of the severity of recurrent acute pancreatitis could expedite more aggres- sive efforts to avoid future morbidity and mortality.

Goals of this investigation

The objective of this study is to compare the differences in patients with first-time acute pancreatitis versus recurrent acute pancreatitis.


Study design and setting

This study is a retrospective chart review at a single academic urban emergency department from 2012 to 2016. This study was approved by the Institutional Review Board with a waiver of consent from all partic- ipants. For all subjects, data abstraction was performed from electronic chart review (Cerner) Charts were reviewed by a minimum of two 0735-6757/(C) 2018

M. Boumezrag et al. / American Journal of Emergency Medicine 37 (2019) 304307 305

trained abstractors using structured data collection sheets and discrep- ancies were resolved by a third abstractor. CT Scans were graded using the Modified CT Severity Index (MCTSI) which grades acute pancreatitis by the presence of inflammation, fluid accumulation, necrosis or extra- pancreatic findings [7].

Selection of participants

Eligible patients were screened through a query of the Cerner electronic medical record for patients with a final ED diagnosis of acute pancreatitis (ICD-9 577.0) After initial screening, a chart re- view was performed to confirm that patients met inclusion criteria including (1) clinical symptoms of pancreatitis (a chief complaint that listed symptoms such as abdominal pain, nausea or vomiting),

(2) age greater than or equal to 18 years, (3) an abdominal CT scan within 24 h of triage and exclusion criteria including Traumatic causes or pregnancy.

Data abstraction and quality assessment

Data was recorded in structured data collection sheets which in- cluded the following data elements: demographics, history of acute pancreatitis, history of recent alcohol use or alcohol abuse, opiate administration, white blood cell count and CT scan findings. Patients were categorized as having an episode of recurrent acute pancreatitis if a history of pancreatitis was documented anywhere in their electronic medical record. Patients were classified as first-time acute pancreatitis if the pancreatitis was not documented anywhere in their medical record. CT findings were graded using the Modified CT Severity Index (MCTSI), which was developed by Balthazar and colleagues to distinguish mild, moderate, and Severe forms of pancreatitis [7,8]. As seen in Fig. 1, this

index gives a score out of 10 indicating the amount of pancreatic inflam- mation and necrosis and is based on the presence of inflammation, fluid accumulation, necrosis, or extra-pancreatic findings. Scores of 0-2, 4-6, and 8-10 correlate to mild, moderate, and severe pancreatitis, respec- tively. Patients were considered to have significant comorbidity if they had any of the following: Chronic renal insufficiency/failure, liver cirrho- sis, congestive heart failure, immunocompromised, cerebrovascular disease or cancer [9].

Data analysis

Continuous and discrete data were summarized using mean or median (standard deviation or variance) and frequency (percentage), respectively. t-Test was used to compare normally distributed continu- ous variables, and Wilcoxon Rank Sum for non-normally distributed continuous variables. Chi-square/Fisher’s exact test was performed to compare discrete variables. An alpha of 0.05 was used as the cutoff for significance. Recurrent pancreatitis was also modeled as a function of selected patient and clinical characteristics using logistic regression. Basic demographic variables were included including age, sex and BMI, as well as clinical and treatment characteristics that were hypoth- esized to be related to recurrent pancreatitis. All analyses were per- formed using SAS 9.4 (SAS Institute, Cary, NC)

Outcome measures

The primary outcome measured was the severity of pancreatitis using the MCTSI in first time vs. recurrent pancreatitis. Secondary out- comes included clinical characteristics, ICU admissions, total hospital length of stay, and amount of opiate received.

Fig. 1. Modified CTSI pancreatic inflammation 0: normal pancreas 2: intrinsic pancreatic abnormalities with or without Inflammatory changes in peri-pancreatic fat 4: pancreatic or peri- pancreatic fluid collection or peri-pancreatic fat necrosis pancreatic necrosis 0: none 2: 30% or less 4: N30% Extra-pancreatic complications 2: one or more of pleural effusion, ascites, vascular complications, parenchymal complications and/or gastrointestinal involvement Total score Total points are given out of 10 to determine the grade of pancreatitis and aid treatment: 0-2: mild 4-6: moderate 8-10: severe *Scores are generated by estimating pancreatic inflammation and necrosis to give a score out of 10.

306 M. Boumezrag et al. / American Journal of Emergency Medicine 37 (2019) 304307


Characteristics of study subjects

Table 2

Predictors of recurrent pancreatitis

Odds ratio 95% confidence interval

In total, 250 patients were included in the study. Of these, 85 patients presented with recurrent acute pancreatitis and 165 patients presented with first-time acute pancreatitis. The median ages for first time and recurrent pancreatitis were 55 and 53, respectively. Females represented 37% (n = 76) of patients in the first-time pancreatitis group and 30% (n = 32) in the recurrent pancreatitis group. Patients with first-time pancreatitis reported a mean triage pain score of 8 while those with recurrent pancreatitis had a mean triage score of 9 (p = 0.002). The ICU disposition was 19% (n = 30) for first-time pan- creatitis and 8% (n = 7) for recurrent pancreatitis (p = 0.03). Median initial lipase was 981 units/L for first-time pancreatitis and 579 units/L for recurrent pancreatitis (p = 0.0002) (Table 1).

We calculated the mean MCTSI score in both groups and found a significantly higher rate of severity in recurrent acute pancreatitis versus first-time acute pancreatitis (2.09 vs. 1.43, p b 0.05) (Fig. 1) 75% (n = 123) of the patients with first-time acute pancreatitis received opiates while 96% (n = 82) of recurrent pancreatitis patients received opiates (p b 0.0001). The quantity of intravenous opiates received, measured by morphine mg equivalents (MME), was signifi- cantly higher in the recurrent vs. first time patients (10 MME vs. 5 MME, p b 0.05).

When modeling recurrent pancreatitis as a function of selected pa- tient and clinical characteristics using logistic regression (Table 2), it was found that intravenous (IV) opiates were more likely to be used in the patients with recurrent disease but IV fluid was less likely to be used. This suggests that clinicians are more focused on pain control with recurrent acute episodes and aggressive resuscitation with first- time episodes.


In our study, ED patients with recurrent acute pancreatitis were more likely to present with more significant findings on abdominal CT than those experiencing a first-time episode of acute pancreatitis. This study underscores the progressive nature of acute pancreatitis and

Table 1

ED patients with acute pancreatitis

Body mass index (BMI) 1.00 (0.95, 1.04)

Age (years) 1.00 (0.97, 1.02)

Female vs. male 0.64 (0.34, 1.20)

Triage temperature 0.98 (0.47, 2.07)

Triage systolic blood pressure 1.00 (0.99, 1.01)

Initial creatinine mg/dL 0.97 (0.82, 1.16)

Triage heartrate (bpm) 1.00 (0.98, 1.06)

Significant co-morbidity vs. none 1.08 (0.50, 2.33)

alcohol positive vs. negative 0.73 (0.39, 1.36)

IV opiates (morphine mg equivalents) 1.09 (1.04, 1.13)

Liters IV fluid in first 24 h


(0.61, 0.90)

Total hospital LOS (days)


(0.93, 1.05)

suggests that a history of recurrent episodes should be considered when deciding to perform an ED abdominal CT.

Our study should be viewed in the context of current guidelines that recommend that CT scans should not be performed routinely for acute pancreatitis in the first 48 h because of Delayed presentation of compli- cations such as pseudocysts or necrosis [10]. In this study, a relatively high rate of patients with first-time acute pancreatitis (n = 17, 10%) and with recurrent acute pancreatitis (n = 20, 23%) had a MCTSI score greater than or equal to four within 24 h of ED presentation. An MCTSI of four indicates signs of necrosis, inflammation or extra- pancreatic manifestations. Significant CT findings have the potential to change management including the use of antibiotics, surgical debride- ment or interventional radiology to drain infected pseudocysts.

The current understanding of pancreatitis is that disease progresses from acute to chronic pancreatitis due to accumulated damage from prior attacks [11]. Patients with a higher CT severity index have been shown to have higher mortality and morbidity [12]. In our study, the higher MCTSI scores in recurrent episodes may reflect progression of disease over time.

We did not demonstrate that patients with recurrent attacks also demonstrated greater clinical severity. In fact, patients with first-time acute pancreatitis were admitted to the ICU at a higher rate and higher li- pase levels than patients with recurrent disease. It is interesting to note that none of the commonly used clinical risk assessment scores such as the Atlanta classification, the Ranson’s Criteria, the Acute Physiology and Chronic Health Evaluation (APACHE) score or the Modified Glasgow Acute Pancreatitis Severity Score consider a history of acute pancreatitis

First time N = 165

Recurrent N = 85

p value

to be a predictor of disease severity [13]. In addition, the relative amount of lipase elevation is not known to correlate with disease severity. In fact,

Body mass index (BMI) (median)




patients with recurrent episodes of acute pancreatitis may not mount a li-

Age (years) (median)




pase elevation due to the observation that the pancreas will eventually


Fever (N38 ?C)

Triage pain score (median)










become “burned out” and lose exocrine function.

Our study is significant in that it focuses on characterizing differences

Initial creatinine mg/dL (median)




between recurrent episodes of acute pancreatitis and first-time acute

Triage Heartrate (bpm) (median)




pancreatitis. Since most cases of acute pancreatitis are caused by modifi-

able risk factors such as alcohol abuse or biliary disease, a better under- standing of Disease progression may help clinicians address these risk

WBC N 12,000/mm3 (%)




HCT (%)




Lactate (mmol/L)




Symptom duration b48 h 33.7% 42.4% 0.18

Initial lipase units/L (median)




factors at the time of the first event or sooner. For patients with the

Significant co-morbidity




first episode of acute pancreatitis, providing alcoholism treatment, en-

Alcohol positive




couraging dietary modifications and referring to gall bladder surgery

Received opiates 75% 96% b0.0001 Median IV opiates (morphine mg equivalents) 5 10 b0.05 Median liters IV fluid in first 24 h 3 2 0.10

ICU dispo 19% 8% 0.03

Median total hospital LOS (days)




Median MCTSI




MCTSI 0 (n, %)

68, 41%

24, 28%

may lead to a decrease in recurrence and avoidance of the worsening se- quelae [14]. ED intervention for an acute event has been shown to be an effective means of secondary prevention for other chronic diseases [15].


24 h of ED presentation. However, this requirement differs from current clinical practice, in which not all patients presenting to the ED with

MCTSI 2 (n, %)

81, 49%

41, 48%

Our findings should be considered with the following design fea-

MCTSI 4 (n, %)

13, 8%

13, 15%

tures. First, study participants were required to have a CT scan within

MCTSI 6 (n, %)

4, 2%

6, 7%

MCTSI 8 (n, %)

0, 0%

1, 1%

M. Boumezrag et al. / American Journal of Emergency Medicine 37 (2019) 304307 307

suspected acute pancreatitis receive a CT scan in the first 24 h [10,16]. We may have been more likely to enroll patients in whom radiation ex- posure was of less concern, diagnosis was less clear or clinical severity was higher. Second, our study was limited to a single tertiary care center in the US with easy access to intensivists and advanced imaging which may limit the generalizability. Third, a power analysis was not com- pleted a priori and sample size was chose based upon dates of electronic medical record query. Fourth, while we queried whether patient had a history of alcohol abuse or biliary disease, we did not distinguish if the suspected cause of acute pancreatitis was alcohol-related or gallstones. Finally, our study was a retrospective review which may lead to unknown confounders between the two groups.


In conclusion, ED patients with recurrent acute pancreatitis are more likely to present with a more severe episode of acute pancreatitis than patients presenting with first-time acute pancreatitis as measured by CT scan of the abdomen.


Funding Sources: This research did not receive any specific grant from funding agencies in the public, commercial, or non-for profit sectors.


  1. Vege SS, Yadav D, Chari ST. Pancreatitis. GI epidemiology; 2007 (1).
  2. AU Toouli J, Brooke-Smith M, Bassi C, Carr-Locke D, Telford J, Freeny P, et al. Guide- lines for the management of acute pancreatitis. J Gastroenterol Hepatol 2002;17.
  3. Fagenholz PJ, Fernandez-del Castillo C, Harris NS, Pelletier AJ, Camargo CA. National study of United States emergency department visits for acute pancreatitis, 1993-2003. BMC Emerg Med 2007;7(1).
  4. Ahmed Ali U, Issa Y, Hagenaars JC, et al. Risk of recurrent pancreatitis and progres- sion to chronic pancreatitis after a first episode of acute pancreatitis. Clin Gastroenterol Hepatol 2016;14:738-46.
  5. Fu C-Y, Yeh C-N, Hsu J-T, Jan Y-Y, Hwang T-L. Timing of mortality in severe acute pan- creatitis: experience from 643 patients. World J Gastroenterol: WJG 2007;13(13).
  6. Mortele KJ, Ip IK, Wu BU, Conwell DL, Banks PA, Khorasani R. Acute pancreatitis: im- aging utilization practices in an urban teaching hospital–analysis of trends with as- sessment of independent predictors in correlation with patient outcomes. Radiology 2011;258(1):174-81.
  7. Mortele KJ, Wiesner W, Intriere L, Shankar S, Zou KH, Lakantari BN, et al. A modified CT severity index for evaluating acute pancreatitis: improved correlation with pa- tient outcome. Am J Roentgenol 2004;183:1261-5.
  8. Banks PA, Bollen TL, Dervenis C, et al. Classification of acute pancreatitis–2012: re- vision of Atlanta classification and definitions by international consensus. Gut 2013;62:102-11.
  9. Diederichs C, Klaus Berger DB. The measurement of multiple chronic diseases–a sys- tematic review on existing multimorbidity indices. J Gerontol Ser A 2011;66A: 301-11.
  10. Tenner S, Baillie J, Dewitt J, Vege SS. American College of Gastroenterology guide- lines: management of acute pancreatitis. Am J Gastroenterol 2013;108:1400-15.
  11. Witt H, Apte MV, Keim V, Wilson JS. Chronic pancreatitis: challenges and advances in pathogenesis, genetics, diagnosis, and therapy. Gastroenterology 2007;132: 1557-73.
  12. Balthazar EJ, Robinson DL, Megibow AJ, Ranson JH. Acute pancreatitis: value of CT in establishing prognosis. Radiology 1990;174:331-6.
  13. Cho JH, Kim TN, Chung HH, et al. Comparison of scoring systems in predicting the severity of acute pancreatitis. World J Gastroenterol 2015;21:2387-94.
  14. Etemad B, Whitcomb DC. Chronic pancreatitis: diagnosis, classification, and new ge- netic developments. Gastroenterology 2001;120:682-707.
  15. Claret PG, Bobbia X, Jonquet O, Bousquet J, de La Coussaye JE. Integrated chronic dis- ease management to avoid emergency departments: the MACVIA-LR(R) approach. In- tern Emerg Med 2014;9:875-8.
  16. McNabb-Baltar J, Chang MS, Suleiman SL, Banks PA, de Silva DPS. A time trend anal- ysis of CT and MRI scan imaging in acute pancreatitis patients presenting to US emergency departments. Am J Emerg Med 2018;36(9):1709-10.