Impact of prior antibiotics on infected pancreatic necrosis microbiology in ICU patients: a retrospective cohort study

Abstract Background Recent guidelines advise against prophylactic antibiotics in patients with necrotizing pancreatitis, advocating instead a step-up drainage and necrosectomy strategy with antibiotics as dictated by microbiological findings. However, prompt antibiotic therapy is recommended in patients with sepsis or septic shock, a possible presentation of infected pancreatic necrosis (IPN). Consequently, in many critically ill patients with IPN, pancreatic samples are collected only after broad-spectrum antibiotic therapy initiation. Whether this prior antibiotic exposure alters the microbiological findings is unknown. The main objective was to determine whether prior antibiotic exposure sterilized the samples collected during procedures for suspected IPN in patients admitted to the intensive care unit (ICU) for acute pancreatitis with suspected IPN. We retrospectively studied 56 consecutive ICU patients admitted with suspected IPN. We collected details on the microbiological samples and antimicrobials used. A definite diagnosis of IPN was given when bacteria were identified in pancreatic samples. Results In all, 137 pancreatic samples were collected, including 91 (66.4%) after antibiotic therapy initiation. IPN was confirmed in 48 (86%) patients. The proportion of positive samples was 74 (81.3%) in antibiotic-exposed patients and 32/46 (69.5%) in unexposed patients (p = 0.58). Of the 74 positive samples from exposed patients, 62 (84%) had organisms susceptible to the antibiotics used. One-third of samples contained more than one organism. Among patients with IPN, 37.5% had positive blood cultures. Multidrug- or extensively drug-resistant bacteria were identified at some point in half the patients. Enterobacter cloacae complex was more frequent in the exposed group (p = 0.02), as were Gram-negative anaerobic bacteria (p = 0.03). Conclusion Antibiotic exposure before sampling did not seem to affect culture positivity of pancreatic samples to confirm IPN, but may affect microbiological findings. Our results suggest that, in patients with sepsis and suspected IPN, antibiotics should be started immediately and pancreatic samples obtained as soon as possible thereafter. In other situations, antibiotics can be withheld until the microbiological results of pancreatic samples are available, to ensure accurate targeting of the spectrum to bacterial susceptibility patterns. ClinicalTrials.gov number NCT03253861