Organ Failure and Infection in Necrotizing Pancreatitis: What Are the Predictors of Mortality?
I would like to congratulate Guo et al1 for the article on the role of organ failure and infection in patients with necrotizing pancreatitis. It is indeed a large prospective series, and there is much to learn from their experience. I have a few suggestions and queries regarding this series that I would like to put forward.
I am interested in 321 patients who presented to the hospital directly after the episode of acute pancreatitis and had not been treated earlier. Among these, patients with necrotizing pancreatitis should be analyzed separately for prognostic scoring. This is the group of patients who, I believe, would give us a wealth of information about the management of this difficult disease. There is a need for a score in necrotizing pancreatitis that would involve both clinical and radiological parameters.
I would contribute a significant proportion of the success of this series to the intensivists in intensive care unit (ICU). I would also like to congratulate the ICU team on the management of pulmonary complications. The mortality in patients with pulmonary failure was 26%, which is commendable.
It was clear on reading the article that all patients with infected necrosis needed intervention. However, only 28% of patients with sterile necrosis needed intervention. I believe that the predominant indication for intervention in this group was organ failure. However, it would be good to hear the author's opinion on this matter.
It was stated in the article that 68% in the sterile group and 71% in the infected group had organ failure. Because the mean APACHE II score and the median CT severity index of patients with sterile and infected necrosis is the same, it is difficult to tell in which group sicker patients were, at the time of presentation. If the mortality in the sterile and the infected necrosis group were similar, as the article suggests, then this difference is of less clinical importance and should be abandoned. But infection in necrosis was surely an indication for intervention, though not a predictor of mortality. It was not stated how many of these patients had walled-off pancreatic necrosis. It is now becoming increasingly clear that patients of walled-off pancreatic necrosis have good outcome as the body is well able to wall off the inflammation.2 These patients do well with minimally invasive necrosectomy.
What was the antibiotic protocol used? What was the mean duration that patients with sterile and infected necrosis were on antibiotic treatment? If antibiotic usage was different in these groups, it would signify that patients with sterile and infected necrosis were managed differently. If not, I would believe that there was no clear-cut protocol on the use of antibiotic. What proportion of patients needed antifungal treatment? Was development of fungal sepsis a risk factor for mortality? These data were not presented in the article.
The organ failure (OF) was studied in detail in the article. The article suggests that persistent OF and multiorgan failure in the first week was more in patients with infected necrosis than sterile necrosis. However, a number of aspects regarding OF were ignored. What was the time to organ failure in patients with necrotizing pancreatitis? And what was the time to multiorgan failure? With regard to OF, patients with early mortality (<1 week) should be analyzed separately from patients with late mortality (>1 week), as the cause for mortality for these 2 groups is different.
What was the mean number of intervention? The authors stated that minimally invasive step-up approach was not used. So, most of the patients had percutaneous drain placement or open surgery.