1) Mild acute pancreatitis

Mild AP usually resolves in several days. In the past, oral intake was limited for several days until clinical symptoms improve with normalizing amylase, lipase level. Today, if tolerated, immediate enteral feeding is recommended with avoiding parenteral nutrition [3-7]. Mostly, hydration is enough for patients with mild pancreatitis. Studies demonstrated shorter duration of intravenous fluid, shorter fasting days, and shorter length of hospital stay in the oral feeding group compared to fasting group in mild AP [13-15]. According to initial study on enteral feeding, enteral feeding group showed significant decrease in APACH II score and C-reactive protein (CRP) level during acute phase of pancreatitis, implying that enteral feeding modulates the inflammatory and sepsis response [13]. In real world, patients are still fed in an increasing manner in many centers; clear liquid, low-fat soft diet, low fat solid diet [16]. However, immediate solid diet was found out to be safe (no difference in pain relapse) and can reduce the length of hospital stay [16].

2) Severe acute pancreatitis

Severe AP is systemic inflammatory response syndrome (SIRS), which increases basal metabolic rate and energy consumption leading to a catabolic state [17]. Catabolic stress with nutritional deficiency give negative clinical outcome, especially in malnourished alcoholics. Therefore, nutrition is much more important in severe AP than in mild AP. Guidelines recommend enteral feeding in severe AP, possibly within 48 hours of admission [3-7,18]. A meta-analysis performed subgroup analysis with studies with severe AP, and enteral feeding showed much better outcome in mortality (RR 0.18) compared to that of all patients with AP (RR 0.5) [12]. Other meta-analyses of studies for severe AP have shown clinical benefits of enteral feeding over parenteral feeding, e.g., reducing infectious complications and mortality [19,20]. These results are important and oral feeding should be started as soon as possible. However, in patients with severe AP, conditions often do not allow oral feeding because of ileus, organ failure and other metabolic problems. For those who are unable to tolerate oral intake, early nutritional support should be considered within the first 24-72 hours by nasogastric or nasojejunal tube [18]. It is quite clear that enteral feeding shows better clinical outcomes compared to parenteral feeding. However, the timing of feeding remains controversial [15]. Bakker et al. [21] conducted a randomized trial comparing early nasoenteric tube feeding and an oral diet at 72 hours after presentation with severe AP. Patients were randomly assigned either to early group (nasoenteric tube feeding within 24 hours) or to on-demand group (diet initiated 72 hours after presentation), and early group did not show the superiority in reducing infection or mortality [21].

3) Nasogastric or nasojejunal tube feeding

Nasojejunal tube may have merits in preventing aspiration pneumonia and resting pancreas, but a meta-analysis was not able to conclude any superiority or inferiority between nasogastric and nasojejunal feeding [22]. Considering the merits of nasogastric tube (easier placement, cheaper cost), nasojejunal tube should be used for patients with gastric outlet obstruction [7,18].

4) Formula and probiotics

ESPEN guidelines recommend standard polymeric diet in patients with AP [23]. Immuno-enhanced formulations, such as formula containing glutamine, omega-3 fatty acids, nucleotides, fibre-enhanced formulations have been tried, but they did not show clear clinical benefits [24]. For prevention of bacterial translocation from the intestinal tract to the blood stream and necrotic tissue, probiotics were tried as it could stabilize intestinal barrier. However, Besselink et al.’s randomized controlled study with probiotic prophylaxis for severe AP did not reduce infectious complications [19]. On the other hand, surgical intervention, bowel ischemia and mortality was higher in probiotics group. Meta-analysis also did not show no benefit of probiotics in AP [25]. Therefore, guidelines are against using probiotics yet [5,7].

1) Prophylactic antibiotic

Guidelines do not recommend routine use of prophylactic antibiotics as they did not decrease in mortality [4-6]. Several RCTs have shown no benefit of prophylactic antibiotics for preventing infection, mortality [28-30]. No difference of mortality was consistently reported by meta-analysis of RCTs [31-33]. Poor penetration to necrosis is thought to be one of the reasons of the result, and routine prophylactic antibiotics may lead to infection by multidrug resistance bacteria or fungus with elevation of medical cost. However, there are some reports favoring prophylactic antibiotics in severe pancreatitis. Japanese guideline recommend using prophylactic wide-spectrum antibiotics for severe AP within 72 hours of onset [6]. Their meta-analysis team focused on timing of antibiotics administration, and trials with early administration (72 hours after onset of symptoms or 48 hours after admission) trials were included. Antibiotics group showed better result in mortality rates (OR 0.48; 95% confidence interval [CI] 0.25-0.94) and infection of pancreatic necrosis (OR 0.55; 95% CI 0.33-0.92) [34]. UK guideline recommend considering prophylactic antibiotics in patients with necrosis more than 30% as the risk of infected necrosis is very small when necrosis is less than 30% [4].

Even though prophylactic antibiotics is not generally recommended, broad spectrum antibiotic should be initiated when infection is suspicious. Image (gas in a pancreatic or peripancreatic collection), laboratory examination (worsening white blood cell [WBC], CRP, procalcitonin), clinical symptom/sign (sustained fever, persistent unwellness, or clinical deterioration) are suggested [18]. However, WBC or CRP should not be sole factor to decide to use antibiotics [27]. As many of these factors can be seen in the setting of SIRS, distinguishing infected necrosis could be difficult based on clinical parameters alone.

2) Which antibiotics should be used?

For selection of antibiotics, penetration to pancreatic tissue as well as necrotic tissue and specific activity against the common bacteria for pancreatic infection should be considered. Major organisms are gastrointestinal Gram-negative bacteria (E. coli, Pseudomonas, Proteus, Klebsiella ). Traditionally, carbapenem and quinolone have been used for acute pancreatitis. The 3rd or higher generation cephalosporine also seems to show good penetration to the pancreatic tissue [35,36].