We thank Drs Chandrasekhara and Ahmad for their interest and thoughtful commentary on our study. “Let the dead be removed rather than buried!” That was the principle behind surgical practice of necrosectomy for necrotizing pancreatitis with convincing visuals of large chunks of surgically removed necrotic material. However, learned surgeons turned the conventional wisdom on its head by astute clinical observations that conservative strategy was better than surgery for sterile necrosis because operative trauma and collateral damage added injury to insult (Am J Surg 1991;161:19–24). Self-skepticism is the ultimate test of scientific scrutiny! However, that was for sterile necrosis, or so we continued to believe. The story of infected necrosis evolved more gradually, and again the credit goes to the surgeons for having realized that the mortality was unacceptably high with open necrosectomy, and with the earlier approach was better. Thus, of late, the contrarian “lesser and later” strategy has been accepted. Infected necrosis is sterile necrosis plus infection. A few series have in fact shown that there was not much difference between sterile and infected necroses in terms of mortality, possibly because infection was never ruled out in most patients with sterile necrosis, and the issue of colonization versus infection was not settled in the case of “infected” necrosis (Am J Surg 1992;163:105–110; Ann Surg, 2000;232:627–629). So, if we treat infection and make so-called infected necrosis sterile, we should, at least theoretically, be able to treat infected necrosis conservatively. Examples of other serious, deep-seated, and life-threatening infections being treated primarily with a conservative approach should strengthen such an argument, necrotizing pneumonia being a case in point. But logical argument is no proof of outcome inquiry. A recent systematic review of available data has indeed shown that patients with infected necrosis could be treated successfully with antibiotics and percutaneous drainage without necrosectomy (Br J Surg 2011;98:18–27). Even after excluding the case series reporting successful conservative treatment in select patients (more by serendipity than design) to avoid publication bias, studies employing primary nonoperative treatment for consecutive patients with infected necrosis had shown the feasibility and success of conservative management (Pancreas 2005;30:195–198). Because surgery is advisable only after 4 weeks, it is prudent to take advantage of the natural reparative processes that contain and liquefy the necrosed tissue, thus rendering it amenable to conservative treatment as borne out by our study. A few issues in our study merit clarification as pointed out by the reviewers. The retrospective nature of our study was only but natural in view of the overwhelming surgical bias in earlier times. Let's leave aside for awhile the issue of comparison of conservative strategy with primary surgery owing to perceived differences in the management sophistication between the 2 time periods. That surgery could be avoided completely in 77% of patients with infected necrosis with a mortality of 28% belies the earlier held belief that the mortality in nonoperatively treated patients approaches 100% (Br J Surg 1993;80:148–154). Of course, improvements in intensive care, a better understanding of the pathophysiologic perturbations, and protocolized management have contributed to the success of the conservative strategy but that only strengthens the case for such an approach. In our study, all consecutive patients were treated with conservative-first approach in the later time period (2003–2008) and those who failed conservative treatment were treated surgically. Thus, there was no selection bias for conservative treatment in that period, although that could have been the case in the first time period when only patients who showed unequivocal improvement or could not undergo surgery were treated conservatively. The greater mortality in the conservative group in our study was most likely from referral bias, because 56% patients had organ failure at the time of admission. Of the 11 deaths in the conservative group, 9 died owing to persistent organ failure, 1 from morbid obesity and pulmonary thromboembolism, and the remaining patient from complications of exploratory laparotomy. Because the critical determinant in severe acute pancreatitis is organ failure, the fate of the 9 patients with persistent organ failure could only be guessed had they been operated. While following the conservative-first approach, it is important to recognize at an appropriate time that some patients will require some form of necrosectomy at a later date. The logical indications for necrosectomy—poorly localized and solid necrotic debris, and a relatively good surgical risk patient not yet improving—require evidence from studies designed specifically to address this particular question. Primary Conservative Therapy in Infected Pancreatic Necrosis: A Viable Treatment Option?GastroenterologyVol. 141Issue 2PreviewGarg PK, Sharma M, Madan K, et al. (Department of Gastroenterology, All India Institute of Medical Sciences, New Delhi, India). Primary conservative treatment results in mortality comparable to surgery in patients with infected pancreatic necrosis. Clin Gastroenterol Hepatol 2010;8:1089–1094.e2. Full-Text PDF
Read full abstract
Dec 1, 2012
Souheil Abdel Nour + 4
Open Access