Residency · Residency · Internal Medicine
Acute Pancreatitis: Severity Assessment and Fluid Management
Diagnosis
Diagnostic Criteria (2 of 3 required -- Revised Atlanta Classification)
The diagnosis of acute pancreatitis requires two of the following three criteria: characteristic abdominal pain (acute onset, severe, epigastric, radiating to the back), serum lipase or amylase at or above three times the upper limit of normal, and characteristic findings on cross-sectional imaging (CT, MRI, or ultrasound). Lipase is preferred over amylase because it is more sensitive and specific and remains elevated longer. Imaging is not required for diagnosis if criteria one and two are both met. CT should be reserved for diagnostic uncertainty or assessment of complications, ideally delayed 72-96 hours to allow necrosis to become apparent.
Etiology -- The First Hospitalization Priority
Gallstones account for approximately 40% of cases; right upper quadrant ultrasound should be obtained in all patients, and an ALT above 150 U/L has an 85% positive predictive value for gallstone etiology. Alcohol is responsible for 25-35% and typically requires more than 5 years of heavy use (over 50 g/day). Hypertriglyceridemia (1-4%) causes pancreatitis when serum triglycerides exceed 1000 mg/dL, and fasting lipids should be checked. Post-ERCP pancreatitis (5-10%) is usually mild and self-limited. Medications implicated include azathioprine, valproic acid, GLP-1 agonists (debated), ACE inhibitors, and didanosine. Autoimmune pancreatitis has two types: type 1 (IgG4-related) and type 2, and should be considered if imaging is atypical. Pancreatic malignancy should be considered in patients over 40 with a first episode and no clear etiology. Other causes include pancreas divisum, sphincter of Oddi dysfunction, and genetic mutations (PRSS1, SPINK1, CFTR). Idiopathic cases account for 15-25% after thorough workup; EUS should be considered to evaluate for occult stones or neoplasm.
Severity Assessment
Revised Atlanta Classification (2012)
| Severity | Definition | Frequency | Mortality |
|---|---|---|---|
| Mild | No organ failure, no local complications | ~80% | <1% |
| Moderately Severe | Transient organ failure (<48h) and/or local complications | ~15% | <5% |
| Severe | Persistent organ failure (>48h) | ~5% | 20-40% |
Mild pancreatitis, representing 80% of cases with mortality below 1%, involves no organ failure and no local complications. Moderately severe pancreatitis includes transient organ failure (lasting less than 48 hours) and/or local complications. Severe pancreatitis is defined by persistent organ failure exceeding 48 hours and carries mortality of 20-40%.
Organ Failure Assessment (Modified Marshall Score)
The Modified Marshall Score assesses respiratory (PaO2/FiO2), cardiovascular (systolic blood pressure with or without fluid/vasopressor response), and renal (creatinine) function. A score of 2 or greater in any system defines organ failure.
Prognostic Scoring Systems
| Scoring System | Components | Timing | Key Threshold |
|---|---|---|---|
| BISAP | BUN >25, Impaired mental status, SIRS, Age >60, Pleural effusion | Admission | ≥3 = increased mortality |
| APACHE II | 12 physiologic variables + age + chronic health | Admission + ongoing | ≥8 = severe disease |
| Ranson | 11 criteria (5 at admission, 6 at 48h) | 0 and 48 hours | ≥3 = severe (largely replaced) |
| CT Severity Index | Inflammation grade + necrosis extent | 72-96 hours | Higher score = worse prognosis |
| HAPS | No rebound/guarding, normal Cr, normal Hct | Admission | All met = predicts mild course |
The Bedside Index for Severity in Acute Pancreatitis (BISAP) incorporates BUN above 25, impaired mental status, SIRS, age above 60, and pleural effusion. A score of 3 or higher indicates increased mortality risk, and it can be calculated at admission. Ranson criteria are complex (11 criteria at 0 and 48 hours) and have largely been replaced by BISAP and APACHE II. APACHE II scores of 8 or higher predict severe disease. The CT Severity Index (Balthazar) is based on grade of inflammation and extent of necrosis, assessed at 72 or more hours. The Harmless Acute Pancreatitis Score (HAPS) uses the absence of rebound/guarding, normal creatinine, and normal hematocrit to predict a mild course.
Single Laboratory Predictors
BUN is the simplest and most reliable predictor: admission BUN above 20 and rising BUN at 24 hours predict mortality (Wu et al., Gastroenterology 2009). Hematocrit above 44% at admission and failure to decrease at 24 hours is associated with necrosis. CRP above 150 mg/L at 48 hours predicts severe disease and is widely available. Procalcitonin may predict infected necrosis.
Fluid Management -- The WATERFALL Era
Historical Context
Previously, aggressive early fluid resuscitation at 250-500 mL/hr was standard, based on the rationale of restoring intravascular volume lost to third-spacing and improving pancreatic perfusion.
WATERFALL Trial (2022)
The WATERFALL trial compared aggressive fluids (20 mL/kg bolus plus 3 mL/kg/hr) to moderate resuscitation (1.5 mL/kg/hr with bolus only if hypovolemic) using Lactated Ringer's. The aggressive group experienced increased fluid overload without improved outcomes, while the moderate group had similar clinical outcomes with fewer complications. The current recommendation is goal-directed moderate fluid resuscitation.
Current Best Practice
Lactated Ringer's is preferred over normal saline because it produces less hyperchloremic acidosis and may have anti-inflammatory benefits. The initial rate should be 1.5 mL/kg/hr with frequent reassessment every 6-8 hours. A bolus of 10 mL/kg is administered if there is evidence of hypovolemia (tachycardia, hypotension, oliguria). Goals of resuscitation include urine output of 0.5-1 mL/kg/hr, decreasing BUN, improving hematocrit, and hemodynamic stability. Over-resuscitation should be avoided as it can cause abdominal compartment syndrome, pulmonary edema, and fluid overload.
Pain Management
Pain control is a priority and should not be delayed. First-line agents include IV hydromorphone or morphine; there is no evidence that morphine worsens pancreatitis via sphincter of Oddi spasm, and this belief is a myth. A multimodal approach incorporating acetaminophen and NSAIDs (ketorolac) is recommended. Patient-controlled analgesia (PCA) can be used for severe pain, and epidural analgesia is an option for refractory cases. Meperidine should be avoided due to seizure risk and no advantage over other opioids.
Nutrition
Early Oral Feeding
Oral feeding should begin as soon as tolerated, within 24 hours, regardless of lipase level or symptom resolution. A low-fat solid diet is as safe as a clear liquid diet, eliminating the need for stepwise diet advancement. Early feeding reduces length of stay and does not increase complications.
Enteral Nutrition (if Oral Not Tolerated)
Nasogastric or nasojejunal tube feeding is preferred over TPN. Enteral nutrition maintains gut barrier integrity, reduces bacterial translocation, and decreases both infection and mortality. It should be started within 72 hours if oral intake is not possible.
Total Parenteral Nutrition (TPN)
TPN is reserved for patients who cannot tolerate enteral feeding after failed attempts. It is associated with increased infection risk, gut atrophy, and longer hospital stays and should not be used routinely.
Management of Complications
Local Complications (Revised Atlanta Classification)
Acute peripancreatic fluid collections develop within 4 weeks, lack a capsule, and usually resolve spontaneously. Pancreatic pseudocysts develop after 4 weeks with encapsulation but no necrotic tissue, and are drained if symptomatic, infected, or enlarging. Acute necrotic collections develop within 4 weeks and contain necrotic tissue; they are observed initially. Walled-off necrosis (WON) represents encapsulated necrotic tissue after 4 weeks and is drained if infected or symptomatic.
Infected Pancreatic Necrosis
Infected necrosis should be suspected when clinical deterioration follows initial improvement (usually during weeks 2-4), with persistent fever, rising white blood cell count, or sepsis. CT showing gas bubbles in a necrotic collection is pathognomonic; fine-needle aspiration is no longer routinely recommended due to high false-negative rates and treatment delays. Treatment involves antibiotics with good pancreatic tissue penetration: carbapenems (imipenem/meropenem) are first-line, with fluoroquinolones plus metronidazole as an alternative. The step-up approach, validated by the PANTER and TENSION trials, starts with percutaneous or endoscopic drainage, proceeds to minimally invasive necrosectomy if there is no improvement, and reserves open necrosectomy as a last resort. This approach reduces organ failure and complications compared to primary open necrosectomy. Endoscopic transluminal drainage and necrosectomy through the stomach or duodenum is preferred when expertise is available.
Sterile Necrosis
Sterile necrosis is managed conservatively. Prophylactic antibiotics are not recommended for sterile necrosis as they do not reduce infection or mortality and increase the risk of fungal infections and resistance.
Specific Etiologic Management
Gallstone Pancreatitis
ERCP with sphincterotomy is indicated for concurrent cholangitis or persistent biliary obstruction but is not routinely needed if the patient is improving without cholangitis. Cholecystectomy should be performed during the same hospitalization for mild pancreatitis, reducing recurrence from 30-60% to less than 5%. The PONCHO trial confirmed that same-admission cholecystectomy for mild gallstone pancreatitis reduced readmissions. For severe pancreatitis, cholecystectomy should be delayed 6 or more weeks until inflammation resolves.
Hypertriglyceridemic Pancreatitis
Insulin infusion at 5-10 U/hr activates lipoprotein lipase to clear triglycerides. NPO status stops chylomicron production. Plasmapheresis may be considered for triglycerides above 2000 or refractory cases, though evidence is limited. Long-term management includes fibrates, omega-3 fatty acids, dietary fat restriction, and diabetes control.
Alcohol-Related Pancreatitis
Alcohol cessation counseling is essential because continued use increases recurrence and progression to chronic pancreatitis. Brief intervention and referral for alcohol use disorder treatment should be provided.
<image> A clinical algorithm for acute pancreatitis management. Start with diagnosis (2 of 3 criteria), then etiology workup (RUQ US, labs). Branch into severity assessment using BISAP and Revised Atlanta Classification (mild, moderate, severe). Show management pathways: fluid resuscitation (WATERFALL trial evidence for moderate approach), pain control, nutrition (early oral feeding), and complication management timeline (acute collections <4 weeks vs. walled-off necrosis >4 weeks). </image>
<image> A timeline diagram showing the natural history of acute pancreatitis complications. Horizontal axis from Day 0 to Week 8+. Show the evolution of local complications: acute peripancreatic fluid collection (first 4 weeks) maturing into pseudocyst, and acute necrotic collection maturing into walled-off necrosis. Indicate the optimal timing for intervention (delay until 4+ weeks when possible), CT assessment (72-96 hours), and the step-up approach for infected necrosis. Mark clinical deterioration clues suggesting infection. </image>
<image> An infographic comparing the step-up approach vs. primary open necrosectomy for infected pancreatic necrosis. Show the step-up pathway: percutaneous/endoscopic drainage first, then minimally invasive necrosectomy if no improvement, then open surgery as last resort. Compare outcomes from PANTER and TENSION trials showing reduced organ failure and complications with the step-up approach. Include icons representing each procedure type. </image>
Clinical Pearls
CT imaging is not needed for diagnosis if the patient has classic pain and lipase at or above three times the upper limit of normal; CT should be saved for when complications are suspected (72-96 hours). The WATERFALL trial changed practice: moderate goal-directed fluid resuscitation (1.5 mL/kg/hr with Lactated Ringer's) is preferred over aggressive flooding, as over-resuscitation causes harm. BUN is the simplest and most reliable predictor of prognosis: admission BUN above 20 and rising BUN at 24 hours are red flags. Patients should be fed early (within 24 hours) with a regular low-fat diet; there is no need for NPO status until pain-free or waiting for lipase to normalize. Prophylactic antibiotics for sterile necrosis do not help and are not recommended. The morphine-sphincter of Oddi myth is not supported by evidence, and effective analgesia should not be withheld. Same-admission cholecystectomy for mild gallstone pancreatitis prevents recurrence and patients should not be discharged with a plan for delayed cholecystectomy in mild cases.
References
- Banks PA, et al. Revised Atlanta Classification of Acute Pancreatitis. Gut. 2013.
- de-Madaria E, et al. WATERFALL Trial: Aggressive vs. Moderate Fluid Resuscitation in Acute Pancreatitis. NEJM. 2022.
- van Santvoort HC, et al. PANTER Trial: Step-Up Approach for Necrotizing Pancreatitis. NEJM. 2010.
- van Brunschot S, et al. TENSION Trial: Endoscopic vs. Surgical Step-Up Approach. Lancet. 2018.
- da Costa DW, et al. PONCHO Trial: Same-Admission Cholecystectomy for Mild Gallstone Pancreatitis. Lancet. 2015.
- Wu BU, et al. Blood Urea Nitrogen in Prediction of Severe Acute Pancreatitis. Gastroenterology. 2009.
- Crockett SD, et al. ACG Clinical Guideline: Acute Pancreatitis. Am J Gastroenterol. 2018.
- IAP/APA Guidelines for Management of Acute Pancreatitis. Pancreatology. 2013.
