Operative vs Non-Operative Management in Sterile Necrotizing Pancreatitis

Background: The clinical management of sterile pancreatic necrosis is still a matter of debate. In this study we analyzed the clinical course and outcome of patients with sterile necrotizing pancreatitis treated surgically versus nonsurgically. Study Design: Between May 1982 and December 1993, 249 patients with necrotizing pancreatitis (NP) entered this study, of which 172 (69 percent) had intraoperatively or fine needle aspiration-proven sterile NP. One hundred seven of 172 patients underwent surgery (S group) with necrosectomy and continuous postoperative closed lavage and 65 of 172 were treated by nonsurgical means (NS group). Results: Median Ranson and admission APACHE II scores were 4.7 (range, 1 to 10) and 11 (range, 1 to 29) in the S group, significantly higher than those in the NS group with 3.0 (range, 0 to 6) (p=0.022) and 8 (range, 1 to 23) (p=0.036). After 48 hours of intensive care treatment, APACHE II scores persisted at 10.5 (range, 1 to 29) in the S group and decreased to 6 (range, 0 to 15) (p=0.013) in the NS patients. Median Creactive protein (CRP) levels on admission were 179 mg/L and 68.5 mg/L (p=0.023), respectively. Within 72 hours, 61 (94 percent) of 65 NS-managed patients responded to intensive care therapy, whereas organ complications persisted or increased and thus led to surgery in the S group. Mortality rates were 13.1 percent in the surgically treated patients and 6.2 percent in the nonsurgically treated patients (p=NS). Conclusions: Most patients with limited and sterile pancreatic necrosis respond to intensive care treatment. Indication for surgery in sterile NP should be based on persisting or advancing organ complications despite intensive care therapy. APACHE II scores and adraission CRP levels represent a helpful tool in decision making for surgical or nonsurgical management of NP.

In doubtful cases and in patients with high perioperative risk, a laparoscopy has been performed and in most cases the lesser sac was opened and inspected. With this procedure, we were able to decrease the number of unnecessary laparotomies significantly. In times of cost efficiency and limited resources new diagnostic measures have to be tested against non-invasive cheaper methods. As stated also by the authors, these new methods have to be evaluated .against conventional investigations in prospective comparative studies. Study Design: Between May 1982 and December 1993, 249 patients with necrotizing pancreatitis (NP) entered this study, of which 172 (69 percent) had intraoperatively or fine needle aspiration-proven sterile NP. One hundred seven of 172 patients underwent surgery (S group) with necrosectomy and continuous postoperative closed lavage and 65 of 172 were treated by nonsurgical means (NS group). Results: Median Ranson and admission APACHE II scores were 4.7 (range, 1 to 10) and 11 (range, 1 to 29) in the S group, significantly higher than those in the NS group with 3.0 (range, 0 to 6) (p=0.022) and 8 (range, 1 to 23) (p=0.036). After 48 hours of intensive care treatment, APACHE II scores persisted at 10.5 (range, 1 to 29) in the S group and decreased to 6 (range, 0 to 15) (p=0.013) in the NS patients. Median Creactive protein (CRP) levels on admission were 179 mglL and 68.5 mg/L (p=0.023), respectively. Within 72 hours, 61 (94 percent) of 65 NS-managed patients responded to intensive care therapy, whereas organ complications persisted or increased and thus led to surgery in the S group. Mortality rates were 13.1 percent in the surgically treated patients and 6.2 percent in the nonsurgically treated patients (p=NS). Conclusions Dismissing for the moment the possibility that surgical debridement of SPN is actually harmful, is there any persuasive evidence that operative intervention in these cases is beneficial? Since the tacit assumption underlying Professor Beger's surgical approach to SPN is that debridement will result in a reduction in mortality, it is reasonable to ask whether the surgical approach does in fact improve mortality risk in these patients. Moreover, because the major stated surgical indication in their series was persistent organ failure, is there any evidence that surgical debridement improves the course of organ insufficiency in patients with SPN? It is axiomatic that meaningful evaluation of any putative therapeutic approach requires comparison to an appropriate control group. Since a population of unoperated patients with equally severe SPN was notably absent from their report, we are forced to turn to natural history information in order to evaluate their approach to management. In collected series of 287 unoperated patients with SPN (4)(5)(6)(7)(8), directly comparable to the Ulm series in that the average extent ofpancreatic necrosis was 54%, single or multiple organ failure existed in 39%, and equivalent indicators of severity were present, the overall mortality rate was found to be 9.7% (Table 2).
When this mortality rate derived from these prospective studies is compared to the retrospective 13% surgical mortality reported by Rau et al., no advantage to surgical intervention in comparable patients is apparent. Moreover, there is little available persuasive evidence that surgical intervention favorably affects either the incidence or the course of organ failure. Smadja and Bismuth were unable to detect any beneficial effects of necrosectomy on preexisting organ failure in thier patients, and in fact concluded that surgical intervention exacerbated organ insufficiency2. Similar negative conclusions regarding surgical amelioration of organ failure have been reached by other workers3'4. From a theoretic standpoint, if organ failure were due to the remote effects of circulating noxious substances released by pancreatic necrosis, then removal of the necrotic tissues should result in a relatively rapid reversal of organ dysfunction, given the reasonable supposition that the half-lives of these serum substances should be short. Yet many studies have failed to demonstrate any conclusive benefit to organ failure following with debridement of SPN. Since continuously elevated APACHE II scores usually reflect persisting organ failure, and the average post operative APACHE II scores in the Ulm patients remained relatively constant for at least 7 days following surgery, it is even difficult to demonstrate any favorable organ effects to debridement in their cases.
Ifthere is no unequivocal survival advantage to surgery in comparable patients, no demonstrable amelioration in existing organ failure, and if the potential exists for actual harm in the form of intestinal fistula formation and surgically induced infection, it is not clear that the case has been made for programmatic surgical debridement in patients with SPN.
Even though the overwhelming majority of patients with SPN do not appear to benefit from necrosectomy, it remains possible that smaller sub-groups of patients might gain from surgical intervention. In our experience, the only patients for whom debridement of SPN is of unquestioned value are those who develop abdominal pain and hyperamylasemia resulting from attempts at oral feeding after 4-6 weeks of non-operative therapy. This clinical configuration occurs in 5-7% ofcases, and is usually due to necrosis-induced changes in the pancreatic ductal system. Accordingly, the principal issue of contention in this debate is not whether surgical debridement should be done inanypatient with SPN, but rather how frequently it is necessary.
It would seem that additional validation is required before we can accept the Ulm proposals that the majority of patients with SPN require operative intervention, or that the necessity for surgery can be predicted by the duration of organ failure, specific serum levels of C-reactive protein, or APACHE II thresholds. Moreover, the increasing amount of natural history data generated from comparable nonoperated patients, has placed the burden of proof placed squarely upon the shoulders ofthose advocating surgery. It is becoming increasingly clear to many that in the near future surgical debridement of sterile necrotizing pancreatitis will become the exception rather than the rule.