• acute pancreatitis

We thank the authors for their interest in our randomised trial.1 2 First, the patient population: The treatment algorithm outlined in our study is different from the TENSION trial and other preceding trials from the Netherlands.3 In patients with large collections (>120 mm in size), the multigate technique was adopted (20% of patient cohort). In patients with persistent or ongoing systemic inflammatory response syndrome (SIRS) after initial drainage, a CT scan was obtained at 72 hours to assess treatment response. If there was <25% decrease in size of the necrotic collection, then additional interventions were undertaken (another 40% of patient cohort): more drainage tracts were created or necrosectomy was performed depending on contents of the necrotic cavity and/or percutaneous drains were placed. In essence, 60% of study cohort required advanced/additional interventions. We believe that necrosectomy was not required, not because the study population constituted a different patient subset, but because a stepwise, systematic approach precluded the need for necrosectomy in many subjects.

Second, infection: As with 10% of the patients in the endoscopy cohort in the TENSION trial in whom an infection could not be proven, we also had negative cultures and there was no air within the walled-off necrosis cavity on imaging in some subjects. This is an unfortunate limitation of any investigation but as the authors point out it was a small patient subset.

Third, the timing of interventions: 60% of patients were American Society of Anesthesiologists (ASA) class III or IV and 37% had preintervention SIRS on the day of intervention compared to more than 90% of patients in the TENSION trial who had ASA class I/II disease at randomisation.3 Although we do not know what the ASA scores were at the time of intervention in the TENSION trial, in our experience most patients are sicker at presentation and as the necrosis becomes more liquefied (and conducive for an intervention) patients are actually clinically better.

Similar to our study, in a recent multicentre trial that compared the lumen-apposing metal and plastic stents in 149 patients with pancreatic fluid collections, there was no significant difference in clinical success or the number of interventions between treatment groups.4 The few constants about necrotising pancreatitis is that it is a morbidly complex disease that requires multidisciplinary care. There are several variations to management and we hope that we have shared some of our perspectives.